diff --git a/.env.example b/.env.example index 06d43070c..bc68456d6 100644 --- a/.env.example +++ b/.env.example @@ -17,4 +17,4 @@ # BRAVE_SEARCH_API_KEY=BSA... # ── Timezone ────────────────────────────── -TZ=Asia/Tokyo +TZ=Asia/Shanghai diff --git a/LICENSE b/LICENSE index 410acae26..b38d9340d 100644 --- a/LICENSE +++ b/LICENSE @@ -19,7 +19,3 @@ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. - ---- - -PicoClaw is heavily inspired by and based on [nanobot](https://github.com/HKUDS/nanobot) by HKUDS. diff --git a/README.fr.md b/README.fr.md index e537fc13a..320aa9e22 100644 --- a/README.fr.md +++ b/README.fr.md @@ -827,7 +827,7 @@ Le sous-agent a accès aux outils (message, web_search, etc.) et peut communique ### Fournisseurs > [!NOTE] -> Groq fournit la transcription vocale gratuite via Whisper. Si configuré, les messages vocaux Telegram seront automatiquement transcrits. +> Groq fournit la transcription vocale gratuite via Whisper. Si configuré, les messages audio de n'importe quel canal seront automatiquement transcrits au niveau de l'agent. | Fournisseur | Utilisation | Obtenir une Clé API | | ------------------------ | ---------------------------------------- | ------------------------------------------------------ | diff --git a/README.ja.md b/README.ja.md index 20ad5033b..ea6bc7e72 100644 --- a/README.ja.md +++ b/README.ja.md @@ -785,7 +785,7 @@ HEARTBEAT_OK 応答 ユーザーが直接結果を受け取る ### プロバイダー > [!NOTE] -> Groq は Whisper による無料の音声文字起こしを提供しています。設定すると、Telegram の音声メッセージが自動的に文字起こしされます。 +> Groq は Whisper による無料の音声文字起こしを提供しています。設定すると、あらゆるチャンネルからの音声メッセージがエージェントレベルで自動的に文字起こしされます。 | プロバイダー | 用途 | API キー取得先 | | --- | --- | --- | diff --git a/README.md b/README.md index c5b38e222..c1ef72141 100644 --- a/README.md +++ b/README.md @@ -54,7 +54,7 @@ ## 📢 News -2026-02-16 🎉 PicoClaw hit 12K stars in one week! Thank you all for your support! PicoClaw is growing faster than we ever imagined. Given the high volume of PRs, we urgently need community maintainers. Our volunteer roles and roadmap are officially posted [here](docs/ROADMAP.md) —we can’t wait to have you on board! +2026-02-16 🎉 PicoClaw hit 12K stars in one week! Thank you all for your support! PicoClaw is growing faster than we ever imagined. Given the high volume of PRs, we urgently need community maintainers. Our volunteer roles and roadmap are officially posted [here](ROADMAP.md) —we can’t wait to have you on board! 2026-02-13 🎉 PicoClaw hit 5000 stars in 4days! Thank you for the community! There are so many PRs & issues coming in (during Chinese New Year holidays), we are finalizing the Project Roadmap and setting up the Developer Group to accelerate PicoClaw's development. 🚀 Call to Action: Please submit your feature requests in GitHub Discussions. We will review and prioritize them during our upcoming weekly meeting. @@ -911,7 +911,7 @@ The subagent has access to tools (message, web_search, etc.) and can communicate ### Providers > [!NOTE] -> Groq provides free voice transcription via Whisper. If configured, Telegram voice messages will be automatically transcribed. +> Groq provides free voice transcription via Whisper. If configured, audio messages from any channel will be automatically transcribed at the agent level. | Provider | Purpose | Get API Key | | -------------------------- | --------------------------------------- | -------------------------------------------------------------------- | diff --git a/README.pt-br.md b/README.pt-br.md index bfe655770..67ce9e0d3 100644 --- a/README.pt-br.md +++ b/README.pt-br.md @@ -823,7 +823,7 @@ O subagente tem acesso às ferramentas (message, web_search, etc.) e pode se com ### Provedores > [!NOTE] -> O Groq fornece transcrição de voz gratuita via Whisper. Se configurado, mensagens de voz do Telegram serão automaticamente transcritas. +> O Groq fornece transcrição de voz gratuita via Whisper. Se configurado, mensagens de áudio de qualquer canal serão automaticamente transcritas no nível do agente. | Provedor | Finalidade | Obter API Key | | --- | --- | --- | diff --git a/README.vi.md b/README.vi.md index b30659614..5755896ed 100644 --- a/README.vi.md +++ b/README.vi.md @@ -795,7 +795,7 @@ Subagent có quyền truy cập các công cụ (message, web_search, v.v.) và ### Nhà cung cấp (Providers) > [!NOTE] -> Groq cung cấp dịch vụ chuyển giọng nói thành văn bản miễn phí qua Whisper. Nếu đã cấu hình Groq, tin nhắn thoại trên Telegram sẽ được tự động chuyển thành văn bản. +> Groq cung cấp dịch vụ chuyển giọng nói thành văn bản miễn phí qua Whisper. Nếu đã cấu hình Groq, tin nhắn âm thanh từ bất kỳ kênh nào sẽ được tự động chuyển thành văn bản ở cấp độ agent. | Nhà cung cấp | Mục đích | Lấy API Key | | --- | --- | --- | diff --git a/README.zh.md b/README.zh.md index db96ba555..bd90173f9 100644 --- a/README.zh.md +++ b/README.zh.md @@ -459,7 +459,7 @@ Agent 读取 HEARTBEAT.md ### 提供商 (Providers) > [!NOTE] -> Groq 通过 Whisper 提供免费的语音转录。如果配置了 Groq，Telegram 语音消息将被自动转录为文字。 +> Groq 通过 Whisper 提供免费的语音转录。如果配置了 Groq，任意渠道的音频消息都将在 Agent 层面自动转录为文字。 | 提供商 | 用途 | 获取 API Key | | -------------------- | ---------------------------- | -------------------------------------------------------------------- | diff --git a/assets/wechat.png b/assets/wechat.png index 1c0b88295..32998c122 100644 Binary files a/assets/wechat.png and b/assets/wechat.png differ diff --git a/cmd/picoclaw/internal/gateway/helpers.go b/cmd/picoclaw/internal/gateway/helpers.go index 747f7d44e..174f5db62 100644 --- a/cmd/picoclaw/internal/gateway/helpers.go +++ b/cmd/picoclaw/internal/gateway/helpers.go @@ -36,6 +36,7 @@ import ( "github.com/sipeed/picoclaw/pkg/providers" "github.com/sipeed/picoclaw/pkg/state" "github.com/sipeed/picoclaw/pkg/tools" + "github.com/sipeed/picoclaw/pkg/voice" ) func gatewayCmd(debug bool) error { @@ -134,6 +135,12 @@ func gatewayCmd(debug bool) error { agentLoop.SetChannelManager(channelManager) agentLoop.SetMediaStore(mediaStore) + // Wire up voice transcription if a supported provider is configured. + if transcriber := voice.DetectTranscriber(cfg); transcriber != nil { + agentLoop.SetTranscriber(transcriber) + logger.InfoCF("voice", "Transcription enabled (agent-level)", map[string]any{"provider": transcriber.Name()}) + } + enabledChannels := channelManager.GetEnabledChannels() if len(enabledChannels) > 0 { fmt.Printf("✓ Channels enabled: %s\n", enabledChannels) @@ -223,19 +230,25 @@ func setupCronTool( // Create cron service cronService := cron.NewCronService(cronStorePath, nil) - // Create and register CronTool - cronTool, err := tools.NewCronTool(cronService, agentLoop, msgBus, workspace, restrict, execTimeout, cfg) - if err != nil { - log.Fatalf("Critical error during CronTool initialization: %v", err) + // Create and register CronTool if enabled + var cronTool *tools.CronTool + if cfg.Tools.IsToolEnabled("cron") { + var err error + cronTool, err = tools.NewCronTool(cronService, agentLoop, msgBus, workspace, restrict, execTimeout, cfg) + if err != nil { + log.Fatalf("Critical error during CronTool initialization: %v", err) + } + + agentLoop.RegisterTool(cronTool) } - agentLoop.RegisterTool(cronTool) - - // Set the onJob handler - cronService.SetOnJob(func(job *cron.CronJob) (string, error) { - result := cronTool.ExecuteJob(context.Background(), job) - return result, nil - }) + // Set onJob handler + if cronTool != nil { + cronService.SetOnJob(func(job *cron.CronJob) (string, error) { + result := cronTool.ExecuteJob(context.Background(), job) + return result, nil + }) + } return cronService } diff --git a/cmd/picoclaw/internal/skills/install.go b/cmd/picoclaw/internal/skills/install.go index a30f68632..78bc421db 100644 --- a/cmd/picoclaw/internal/skills/install.go +++ b/cmd/picoclaw/internal/skills/install.go @@ -21,8 +21,8 @@ picoclaw skills install --registry clawhub github `, Args: func(cmd *cobra.Command, args []string) error { if registry != "" { - if len(args) != 2 { - return fmt.Errorf("when --registry is set, exactly 2 arguments are required: ") + if len(args) != 1 { + return fmt.Errorf("when --registry is set, exactly 1 argument is required: ") } return nil } @@ -45,7 +45,7 @@ picoclaw skills install --registry clawhub github return err } - return skillsInstallFromRegistry(cfg, args[0], args[1]) + return skillsInstallFromRegistry(cfg, registry, args[0]) } return skillsInstallCmd(installer, args[0]) diff --git a/cmd/picoclaw/internal/skills/install_test.go b/cmd/picoclaw/internal/skills/install_test.go index 97787a986..6b362822d 100644 --- a/cmd/picoclaw/internal/skills/install_test.go +++ b/cmd/picoclaw/internal/skills/install_test.go @@ -26,3 +26,72 @@ func TestNewInstallSubcommand(t *testing.T) { assert.Len(t, cmd.Aliases, 0) } + +func TestInstallCommandArgs(t *testing.T) { + tests := []struct { + name string + args []string + registry string + expectError bool + errorMsg string + }{ + { + name: "no registry, one arg", + args: []string{"sipeed/picoclaw-skills/weather"}, + registry: "", + expectError: false, + }, + { + name: "no registry, no args", + args: []string{}, + registry: "", + expectError: true, + errorMsg: "exactly 1 argument is required: ", + }, + { + name: "no registry, too many args", + args: []string{"arg1", "arg2"}, + registry: "", + expectError: true, + errorMsg: "exactly 1 argument is required: ", + }, + { + name: "with registry, one arg", + args: []string{"weather-skill"}, + registry: "clawhub", + expectError: false, + }, + { + name: "with registry, no args", + args: []string{}, + registry: "clawhub", + expectError: true, + errorMsg: "when --registry is set, exactly 1 argument is required: ", + }, + { + name: "with registry, too many args", + args: []string{"arg1", "arg2"}, + registry: "clawhub", + expectError: true, + errorMsg: "when --registry is set, exactly 1 argument is required: ", + }, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + cmd := newInstallCommand(nil) + + if tt.registry != "" { + require.NoError(t, cmd.Flags().Set("registry", tt.registry)) + } + + err := cmd.Args(cmd, tt.args) + if tt.expectError { + require.Error(t, err) + assert.Equal(t, tt.errorMsg, err.Error()) + } else { + require.NoError(t, err) + } + }) + } +} diff --git a/config/config.example.json b/config/config.example.json index 3c84cfa9f..ef1bf3eda 100644 --- a/config/config.example.json +++ b/config/config.example.json @@ -6,7 +6,9 @@ "model_name": "gpt4", "max_tokens": 8192, "temperature": 0.7, - "max_tool_iterations": 20 + "max_tool_iterations": 20, + "summarize_message_threshold": 20, + "summarize_token_percent": 75 } }, "model_list": [ @@ -20,7 +22,8 @@ "model_name": "claude-sonnet-4.6", "model": "anthropic/claude-sonnet-4.6", "api_key": "sk-ant-your-key", - "api_base": "https://api.anthropic.com/v1" + "api_base": "https://api.anthropic.com/v1", + "thinking_level": "high" }, { "model_name": "gemini", @@ -59,6 +62,7 @@ "discord": { "enabled": false, "token": "YOUR_DISCORD_BOT_TOKEN", + "proxy": "", "allow_from": [], "group_trigger": { "mention_only": false @@ -221,27 +225,48 @@ "mistral": { "api_key": "", "api_base": "https://api.mistral.ai/v1" + }, + "avian": { + "api_key": "", + "api_base": "https://api.avian.io/v1" } }, "tools": { + "allow_read_paths": null, + "allow_write_paths": null, "web": { + "enabled": true, "brave": { "enabled": false, "api_key": "YOUR_BRAVE_API_KEY", "max_results": 5 }, + "tavily": { + "enabled": false, + "api_key": "", + "base_url": "", + "max_results": 0 + }, "duckduckgo": { "enabled": true, "max_results": 5 }, "perplexity": { "enabled": false, - "api_key": "pplx-xxx", + "api_key": "", "max_results": 5 }, - "proxy": "" + "glm_search": { + "enabled": false, + "api_key": "", + "base_url": "https://open.bigmodel.cn/api/paas/v4/web_search", + "search_engine": "search_std", + "max_results": 5 + }, + "fetch_limit_bytes": 10485760 }, "cron": { + "enabled": true, "exec_timeout_minutes": 5 }, "mcp": { @@ -310,19 +335,75 @@ } }, "exec": { - "enable_deny_patterns": false, - "custom_deny_patterns": [] + "enabled": true, + "enable_deny_patterns": true, + "custom_deny_patterns": null, + "custom_allow_patterns": null }, "skills": { + "enabled": true, "registries": { "clawhub": { "enabled": true, "base_url": "https://clawhub.ai", - "search_path": "/api/v1/search", - "skills_path": "/api/v1/skills", - "download_path": "/api/v1/download" + "auth_token": "", + "search_path": "", + "skills_path": "", + "download_path": "", + "timeout": 0, + "max_zip_size": 0, + "max_response_size": 0 } + }, + "max_concurrent_searches": 2, + "search_cache": { + "max_size": 50, + "ttl_seconds": 300 } + }, + "media_cleanup": { + "enabled": true, + "max_age_minutes": 30, + "interval_minutes": 5 + }, + "append_file": { + "enabled": true + }, + "edit_file": { + "enabled": true + }, + "find_skills": { + "enabled": true + }, + "i2c": { + "enabled": false + }, + "install_skill": { + "enabled": true + }, + "list_dir": { + "enabled": true + }, + "message": { + "enabled": true + }, + "read_file": { + "enabled": true + }, + "spawn": { + "enabled": true + }, + "spi": { + "enabled": false + }, + "subagent": { + "enabled": true + }, + "web_fetch": { + "enabled": true + }, + "write_file": { + "enabled": true } }, "heartbeat": { @@ -337,4 +418,4 @@ "host": "127.0.0.1", "port": 18790 } -} \ No newline at end of file +} diff --git a/pkg/agent/instance.go b/pkg/agent/instance.go index ed438059f..e14acf06d 100644 --- a/pkg/agent/instance.go +++ b/pkg/agent/instance.go @@ -18,22 +18,25 @@ import ( // AgentInstance represents a fully configured agent with its own workspace, // session manager, context builder, and tool registry. type AgentInstance struct { - ID string - Name string - Model string - Fallbacks []string - Workspace string - MaxIterations int - MaxTokens int - Temperature float64 - ContextWindow int - Provider providers.LLMProvider - Sessions *session.SessionManager - ContextBuilder *ContextBuilder - Tools *tools.ToolRegistry - Subagents *config.SubagentsConfig - SkillsFilter []string - Candidates []providers.FallbackCandidate + ID string + Name string + Model string + Fallbacks []string + Workspace string + MaxIterations int + MaxTokens int + Temperature float64 + ThinkingLevel ThinkingLevel + ContextWindow int + SummarizeMessageThreshold int + SummarizeTokenPercent int + Provider providers.LLMProvider + Sessions *session.SessionManager + ContextBuilder *ContextBuilder + Tools *tools.ToolRegistry + Subagents *config.SubagentsConfig + SkillsFilter []string + Candidates []providers.FallbackCandidate } // NewAgentInstance creates an agent instance from config. @@ -57,17 +60,30 @@ func NewAgentInstance( allowWritePaths := compilePatterns(cfg.Tools.AllowWritePaths) toolsRegistry := tools.NewToolRegistry() - toolsRegistry.Register(tools.NewReadFileTool(workspace, readRestrict, allowReadPaths)) - toolsRegistry.Register(tools.NewWriteFileTool(workspace, restrict, allowWritePaths)) - toolsRegistry.Register(tools.NewListDirTool(workspace, readRestrict, allowReadPaths)) - execTool, err := tools.NewExecToolWithConfig(workspace, restrict, cfg) - if err != nil { - log.Fatalf("Critical error: unable to initialize exec tool: %v", err) - } - toolsRegistry.Register(execTool) - toolsRegistry.Register(tools.NewEditFileTool(workspace, restrict, allowWritePaths)) - toolsRegistry.Register(tools.NewAppendFileTool(workspace, restrict, allowWritePaths)) + if cfg.Tools.IsToolEnabled("read_file") { + toolsRegistry.Register(tools.NewReadFileTool(workspace, readRestrict, allowReadPaths)) + } + if cfg.Tools.IsToolEnabled("write_file") { + toolsRegistry.Register(tools.NewWriteFileTool(workspace, restrict, allowWritePaths)) + } + if cfg.Tools.IsToolEnabled("list_dir") { + toolsRegistry.Register(tools.NewListDirTool(workspace, readRestrict, allowReadPaths)) + } + if cfg.Tools.IsToolEnabled("exec") { + execTool, err := tools.NewExecToolWithConfig(workspace, restrict, cfg) + if err != nil { + log.Fatalf("Critical error: unable to initialize exec tool: %v", err) + } + toolsRegistry.Register(execTool) + } + + if cfg.Tools.IsToolEnabled("edit_file") { + toolsRegistry.Register(tools.NewEditFileTool(workspace, restrict, allowWritePaths)) + } + if cfg.Tools.IsToolEnabled("append_file") { + toolsRegistry.Register(tools.NewAppendFileTool(workspace, restrict, allowWritePaths)) + } sessionsDir := filepath.Join(workspace, "sessions") sessionsManager := session.NewSessionManager(sessionsDir) @@ -101,6 +117,22 @@ func NewAgentInstance( temperature = *defaults.Temperature } + var thinkingLevelStr string + if mc, err := cfg.GetModelConfig(model); err == nil { + thinkingLevelStr = mc.ThinkingLevel + } + thinkingLevel := parseThinkingLevel(thinkingLevelStr) + + summarizeMessageThreshold := defaults.SummarizeMessageThreshold + if summarizeMessageThreshold == 0 { + summarizeMessageThreshold = 20 + } + + summarizeTokenPercent := defaults.SummarizeTokenPercent + if summarizeTokenPercent == 0 { + summarizeTokenPercent = 75 + } + // Resolve fallback candidates modelCfg := providers.ModelConfig{ Primary: model, @@ -149,22 +181,25 @@ func NewAgentInstance( candidates := providers.ResolveCandidatesWithLookup(modelCfg, defaults.Provider, resolveFromModelList) return &AgentInstance{ - ID: agentID, - Name: agentName, - Model: model, - Fallbacks: fallbacks, - Workspace: workspace, - MaxIterations: maxIter, - MaxTokens: maxTokens, - Temperature: temperature, - ContextWindow: maxTokens, - Provider: provider, - Sessions: sessionsManager, - ContextBuilder: contextBuilder, - Tools: toolsRegistry, - Subagents: subagents, - SkillsFilter: skillsFilter, - Candidates: candidates, + ID: agentID, + Name: agentName, + Model: model, + Fallbacks: fallbacks, + Workspace: workspace, + MaxIterations: maxIter, + MaxTokens: maxTokens, + Temperature: temperature, + ThinkingLevel: thinkingLevel, + ContextWindow: maxTokens, + SummarizeMessageThreshold: summarizeMessageThreshold, + SummarizeTokenPercent: summarizeTokenPercent, + Provider: provider, + Sessions: sessionsManager, + ContextBuilder: contextBuilder, + Tools: toolsRegistry, + Subagents: subagents, + SkillsFilter: skillsFilter, + Candidates: candidates, } } diff --git a/pkg/agent/loop.go b/pkg/agent/loop.go index b803187b1..1ab79f3ca 100644 --- a/pkg/agent/loop.go +++ b/pkg/agent/loop.go @@ -12,6 +12,7 @@ import ( "errors" "fmt" "path/filepath" + "regexp" "strings" "sync" "sync/atomic" @@ -31,6 +32,7 @@ import ( "github.com/sipeed/picoclaw/pkg/state" "github.com/sipeed/picoclaw/pkg/tools" "github.com/sipeed/picoclaw/pkg/utils" + "github.com/sipeed/picoclaw/pkg/voice" ) type AgentLoop struct { @@ -43,6 +45,7 @@ type AgentLoop struct { fallback *providers.FallbackChain channelManager *channels.Manager mediaStore media.MediaStore + transcriber voice.Transcriber } // processOptions configures how a message is processed @@ -105,71 +108,102 @@ func registerSharedTools( } // Web tools - searchTool, err := tools.NewWebSearchTool(tools.WebSearchToolOptions{ - BraveAPIKey: cfg.Tools.Web.Brave.APIKey, - BraveMaxResults: cfg.Tools.Web.Brave.MaxResults, - BraveEnabled: cfg.Tools.Web.Brave.Enabled, - TavilyAPIKey: cfg.Tools.Web.Tavily.APIKey, - TavilyBaseURL: cfg.Tools.Web.Tavily.BaseURL, - TavilyMaxResults: cfg.Tools.Web.Tavily.MaxResults, - TavilyEnabled: cfg.Tools.Web.Tavily.Enabled, - DuckDuckGoMaxResults: cfg.Tools.Web.DuckDuckGo.MaxResults, - DuckDuckGoEnabled: cfg.Tools.Web.DuckDuckGo.Enabled, - PerplexityAPIKey: cfg.Tools.Web.Perplexity.APIKey, - PerplexityMaxResults: cfg.Tools.Web.Perplexity.MaxResults, - PerplexityEnabled: cfg.Tools.Web.Perplexity.Enabled, - Proxy: cfg.Tools.Web.Proxy, - }) - if err != nil { - logger.ErrorCF("agent", "Failed to create web search tool", map[string]any{"error": err.Error()}) - } else if searchTool != nil { - agent.Tools.Register(searchTool) + if cfg.Tools.IsToolEnabled("web") { + searchTool, err := tools.NewWebSearchTool(tools.WebSearchToolOptions{ + BraveAPIKey: cfg.Tools.Web.Brave.APIKey, + BraveMaxResults: cfg.Tools.Web.Brave.MaxResults, + BraveEnabled: cfg.Tools.Web.Brave.Enabled, + TavilyAPIKey: cfg.Tools.Web.Tavily.APIKey, + TavilyBaseURL: cfg.Tools.Web.Tavily.BaseURL, + TavilyMaxResults: cfg.Tools.Web.Tavily.MaxResults, + TavilyEnabled: cfg.Tools.Web.Tavily.Enabled, + DuckDuckGoMaxResults: cfg.Tools.Web.DuckDuckGo.MaxResults, + DuckDuckGoEnabled: cfg.Tools.Web.DuckDuckGo.Enabled, + PerplexityAPIKey: cfg.Tools.Web.Perplexity.APIKey, + PerplexityMaxResults: cfg.Tools.Web.Perplexity.MaxResults, + PerplexityEnabled: cfg.Tools.Web.Perplexity.Enabled, + GLMSearchAPIKey: cfg.Tools.Web.GLMSearch.APIKey, + GLMSearchBaseURL: cfg.Tools.Web.GLMSearch.BaseURL, + GLMSearchEngine: cfg.Tools.Web.GLMSearch.SearchEngine, + GLMSearchMaxResults: cfg.Tools.Web.GLMSearch.MaxResults, + GLMSearchEnabled: cfg.Tools.Web.GLMSearch.Enabled, + Proxy: cfg.Tools.Web.Proxy, + }) + if err != nil { + logger.ErrorCF("agent", "Failed to create web search tool", map[string]any{"error": err.Error()}) + } else if searchTool != nil { + agent.Tools.Register(searchTool) + } } - fetchTool, err := tools.NewWebFetchToolWithProxy(50000, cfg.Tools.Web.Proxy, cfg.Tools.Web.FetchLimitBytes) - if err != nil { - logger.ErrorCF("agent", "Failed to create web fetch tool", map[string]any{"error": err.Error()}) - } else { - agent.Tools.Register(fetchTool) + if cfg.Tools.IsToolEnabled("web_fetch") { + fetchTool, err := tools.NewWebFetchToolWithProxy(50000, cfg.Tools.Web.Proxy, cfg.Tools.Web.FetchLimitBytes) + if err != nil { + logger.ErrorCF("agent", "Failed to create web fetch tool", map[string]any{"error": err.Error()}) + } else { + agent.Tools.Register(fetchTool) + } } // Hardware tools (I2C, SPI) - Linux only, returns error on other platforms - agent.Tools.Register(tools.NewI2CTool()) - agent.Tools.Register(tools.NewSPITool()) + if cfg.Tools.IsToolEnabled("i2c") { + agent.Tools.Register(tools.NewI2CTool()) + } + if cfg.Tools.IsToolEnabled("spi") { + agent.Tools.Register(tools.NewSPITool()) + } // Message tool - messageTool := tools.NewMessageTool() - messageTool.SetSendCallback(func(channel, chatID, content string) error { - pubCtx, pubCancel := context.WithTimeout(context.Background(), 5*time.Second) - defer pubCancel() - return msgBus.PublishOutbound(pubCtx, bus.OutboundMessage{ - Channel: channel, - ChatID: chatID, - Content: content, + if cfg.Tools.IsToolEnabled("message") { + messageTool := tools.NewMessageTool() + messageTool.SetSendCallback(func(channel, chatID, content string) error { + pubCtx, pubCancel := context.WithTimeout(context.Background(), 5*time.Second) + defer pubCancel() + return msgBus.PublishOutbound(pubCtx, bus.OutboundMessage{ + Channel: channel, + ChatID: chatID, + Content: content, + }) }) - }) - agent.Tools.Register(messageTool) + agent.Tools.Register(messageTool) + } // Skill discovery and installation tools - registryMgr := skills.NewRegistryManagerFromConfig(skills.RegistryConfig{ - MaxConcurrentSearches: cfg.Tools.Skills.MaxConcurrentSearches, - ClawHub: skills.ClawHubConfig(cfg.Tools.Skills.Registries.ClawHub), - }) - searchCache := skills.NewSearchCache( - cfg.Tools.Skills.SearchCache.MaxSize, - time.Duration(cfg.Tools.Skills.SearchCache.TTLSeconds)*time.Second, - ) - agent.Tools.Register(tools.NewFindSkillsTool(registryMgr, searchCache)) - agent.Tools.Register(tools.NewInstallSkillTool(registryMgr, agent.Workspace)) + find_skills_enable := cfg.Tools.IsToolEnabled("find_skills") + install_skills_enable := cfg.Tools.IsToolEnabled("install_skill") + if find_skills_enable || install_skills_enable { + registryMgr := skills.NewRegistryManagerFromConfig(skills.RegistryConfig{ + MaxConcurrentSearches: cfg.Tools.Skills.MaxConcurrentSearches, + ClawHub: skills.ClawHubConfig(cfg.Tools.Skills.Registries.ClawHub), + }) + + if find_skills_enable { + searchCache := skills.NewSearchCache( + cfg.Tools.Skills.SearchCache.MaxSize, + time.Duration(cfg.Tools.Skills.SearchCache.TTLSeconds)*time.Second, + ) + agent.Tools.Register(tools.NewFindSkillsTool(registryMgr, searchCache)) + } + + if install_skills_enable { + agent.Tools.Register(tools.NewInstallSkillTool(registryMgr, agent.Workspace)) + } + } // Spawn tool with allowlist checker - subagentManager := tools.NewSubagentManager(provider, agent.Model, agent.Workspace, msgBus) - subagentManager.SetLLMOptions(agent.MaxTokens, agent.Temperature) - spawnTool := tools.NewSpawnTool(subagentManager) - currentAgentID := agentID - spawnTool.SetAllowlistChecker(func(targetAgentID string) bool { - return registry.CanSpawnSubagent(currentAgentID, targetAgentID) - }) - agent.Tools.Register(spawnTool) + if cfg.Tools.IsToolEnabled("spawn") { + if cfg.Tools.IsToolEnabled("subagent") { + subagentManager := tools.NewSubagentManager(provider, agent.Model, agent.Workspace, msgBus) + subagentManager.SetLLMOptions(agent.MaxTokens, agent.Temperature) + spawnTool := tools.NewSpawnTool(subagentManager) + currentAgentID := agentID + spawnTool.SetAllowlistChecker(func(targetAgentID string) bool { + return registry.CanSpawnSubagent(currentAgentID, targetAgentID) + }) + agent.Tools.Register(spawnTool) + } else { + logger.WarnCF("agent", "spawn tool requires subagent to be enabled", nil) + } + } } } @@ -177,7 +211,7 @@ func (al *AgentLoop) Run(ctx context.Context) error { al.running.Store(true) // Initialize MCP servers for all agents - if al.cfg.Tools.MCP.Enabled { + if al.cfg.Tools.IsToolEnabled("mcp") { mcpManager := mcp.NewManager() // Ensure MCP connections are cleaned up on exit, regardless of initialization success // This fixes resource leak when LoadFromMCPConfig partially succeeds then fails @@ -219,6 +253,7 @@ func (al *AgentLoop) Run(ctx context.Context) error { if !ok { continue } + mcpTool := tools.NewMCPTool(mcpManager, serverName, tool) agent.Tools.Register(mcpTool) totalRegistrations++ @@ -334,6 +369,64 @@ func (al *AgentLoop) SetMediaStore(s media.MediaStore) { al.mediaStore = s } +// SetTranscriber injects a voice transcriber for agent-level audio transcription. +func (al *AgentLoop) SetTranscriber(t voice.Transcriber) { + al.transcriber = t +} + +var audioAnnotationRe = regexp.MustCompile(`\[(voice|audio)(?::[^\]]*)?\]`) + +// transcribeAudioInMessage resolves audio media refs, transcribes them, and +// replaces audio annotations in msg.Content with the transcribed text. +func (al *AgentLoop) transcribeAudioInMessage(ctx context.Context, msg bus.InboundMessage) bus.InboundMessage { + if al.transcriber == nil || al.mediaStore == nil || len(msg.Media) == 0 { + return msg + } + + // Transcribe each audio media ref in order. + var transcriptions []string + for _, ref := range msg.Media { + path, meta, err := al.mediaStore.ResolveWithMeta(ref) + if err != nil { + logger.WarnCF("voice", "Failed to resolve media ref", map[string]any{"ref": ref, "error": err}) + continue + } + if !utils.IsAudioFile(meta.Filename, meta.ContentType) { + continue + } + result, err := al.transcriber.Transcribe(ctx, path) + if err != nil { + logger.WarnCF("voice", "Transcription failed", map[string]any{"ref": ref, "error": err}) + transcriptions = append(transcriptions, "") + continue + } + transcriptions = append(transcriptions, result.Text) + } + + if len(transcriptions) == 0 { + return msg + } + + // Replace audio annotations sequentially with transcriptions. + idx := 0 + newContent := audioAnnotationRe.ReplaceAllStringFunc(msg.Content, func(match string) string { + if idx >= len(transcriptions) { + return match + } + text := transcriptions[idx] + idx++ + return "[voice: " + text + "]" + }) + + // Append any remaining transcriptions not matched by an annotation. + for ; idx < len(transcriptions); idx++ { + newContent += "\n[voice: " + transcriptions[idx] + "]" + } + + msg.Content = newContent + return msg +} + // inferMediaType determines the media type ("image", "audio", "video", "file") // from a filename and MIME content type. func inferMediaType(filename, contentType string) string { @@ -445,6 +538,8 @@ func (al *AgentLoop) processMessage(ctx context.Context, msg bus.InboundMessage) }, ) + msg = al.transcribeAudioInMessage(ctx, msg) + // Route system messages to processSystemMessage if msg.Channel == "system" { return al.processSystemMessage(ctx, msg) @@ -475,8 +570,8 @@ func (al *AgentLoop) processMessage(ctx context.Context, msg bus.InboundMessage) // Reset message-tool state for this round so we don't skip publishing due to a previous round. if tool, ok := agent.Tools.Get("message"); ok { - if mt, ok := tool.(tools.ContextualTool); ok { - mt.SetContext(msg.Channel, msg.ChatID) + if resetter, ok := tool.(interface{ ResetSentInRound() }); ok { + resetter.ResetSentInRound() } } @@ -591,10 +686,7 @@ func (al *AgentLoop) runAgentLoop( } } - // 1. Update tool contexts - al.updateToolContexts(agent, opts.Channel, opts.ChatID) - - // 2. Build messages (skip history for heartbeat) + // 1. Build messages (skip history for heartbeat) var history []providers.Message var summary string if !opts.NoHistory { @@ -614,10 +706,10 @@ func (al *AgentLoop) runAgentLoop( maxMediaSize := al.cfg.Agents.Defaults.GetMaxMediaSize() messages = resolveMediaRefs(messages, al.mediaStore, maxMediaSize) - // 3. Save user message to session + // 2. Save user message to session agent.Sessions.AddMessage(opts.SessionKey, "user", opts.UserMessage) - // 4. Run LLM iteration loop + // 3. Run LLM iteration loop finalContent, iteration, err := al.runLLMIteration(ctx, agent, messages, opts) if err != nil { return "", err @@ -626,21 +718,21 @@ func (al *AgentLoop) runAgentLoop( // If last tool had ForUser content and we already sent it, we might not need to send final response // This is controlled by the tool's Silent flag and ForUser content - // 5. Handle empty response + // 4. Handle empty response if finalContent == "" { finalContent = opts.DefaultResponse } - // 6. Save final assistant message to session + // 5. Save final assistant message to session agent.Sessions.AddMessage(opts.SessionKey, "assistant", finalContent) agent.Sessions.Save(opts.SessionKey) - // 7. Optional: summarization + // 6. Optional: summarization if opts.EnableSummary { al.maybeSummarize(agent, opts.SessionKey, opts.Channel, opts.ChatID) } - // 8. Optional: send response via bus + // 7. Optional: send response via bus if opts.SendResponse { al.bus.PublishOutbound(ctx, bus.OutboundMessage{ Channel: opts.Channel, @@ -649,7 +741,7 @@ func (al *AgentLoop) runAgentLoop( }) } - // 9. Log response + // 8. Log response responsePreview := utils.Truncate(finalContent, 120) logger.InfoCF("agent", fmt.Sprintf("Response: %s", responsePreview), map[string]any{ @@ -766,23 +858,29 @@ func (al *AgentLoop) runLLMIteration( var response *providers.LLMResponse var err error + llmOpts := map[string]any{ + "max_tokens": agent.MaxTokens, + "temperature": agent.Temperature, + "prompt_cache_key": agent.ID, + } + // parseThinkingLevel guarantees ThinkingOff for empty/unknown values, + // so checking != ThinkingOff is sufficient. + if agent.ThinkingLevel != ThinkingOff { + if tc, ok := agent.Provider.(providers.ThinkingCapable); ok && tc.SupportsThinking() { + llmOpts["thinking_level"] = string(agent.ThinkingLevel) + } else { + logger.WarnCF("agent", "thinking_level is set but current provider does not support it, ignoring", + map[string]any{"agent_id": agent.ID, "thinking_level": string(agent.ThinkingLevel)}) + } + } + callLLM := func() (*providers.LLMResponse, error) { if len(agent.Candidates) > 1 && al.fallback != nil { fbResult, fbErr := al.fallback.Execute( ctx, agent.Candidates, func(ctx context.Context, provider, model string) (*providers.LLMResponse, error) { - return agent.Provider.Chat( - ctx, - messages, - providerToolDefs, - model, - map[string]any{ - "max_tokens": agent.MaxTokens, - "temperature": agent.Temperature, - "prompt_cache_key": agent.ID, - }, - ) + return agent.Provider.Chat(ctx, messages, providerToolDefs, model, llmOpts) }, ) if fbErr != nil { @@ -798,11 +896,7 @@ func (al *AgentLoop) runLLMIteration( } return fbResult.Response, nil } - return agent.Provider.Chat(ctx, messages, providerToolDefs, agent.Model, map[string]any{ - "max_tokens": agent.MaxTokens, - "temperature": agent.Temperature, - "prompt_cache_key": agent.ID, - }) + return agent.Provider.Chat(ctx, messages, providerToolDefs, agent.Model, llmOpts) } // Retry loop for context/token errors @@ -964,62 +1058,76 @@ func (al *AgentLoop) runLLMIteration( // Save assistant message with tool calls to session agent.Sessions.AddFullMessage(opts.SessionKey, assistantMsg) - // Execute tool calls - for _, tc := range normalizedToolCalls { - argsJSON, _ := json.Marshal(tc.Arguments) - argsPreview := utils.Truncate(string(argsJSON), 200) - logger.InfoCF("agent", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), - map[string]any{ - "agent_id": agent.ID, - "tool": tc.Name, - "iteration": iteration, - }) + // Execute tool calls in parallel + type indexedAgentResult struct { + result *tools.ToolResult + tc providers.ToolCall + } - // Create async callback for tools that implement AsyncTool - // NOTE: Following openclaw's design, async tools do NOT send results directly to users. - // Instead, they notify the agent via PublishInbound, and the agent decides - // whether to forward the result to the user (in processSystemMessage). - asyncCallback := func(callbackCtx context.Context, result *tools.ToolResult) { - // Log the async completion but don't send directly to user - // The agent will handle user notification via processSystemMessage - if !result.Silent && result.ForUser != "" { - logger.InfoCF("agent", "Async tool completed, agent will handle notification", - map[string]any{ - "tool": tc.Name, - "content_len": len(result.ForUser), - }) + agentResults := make([]indexedAgentResult, len(normalizedToolCalls)) + var wg sync.WaitGroup + + for i, tc := range normalizedToolCalls { + agentResults[i].tc = tc + + wg.Add(1) + go func(idx int, tc providers.ToolCall) { + defer wg.Done() + + argsJSON, _ := json.Marshal(tc.Arguments) + argsPreview := utils.Truncate(string(argsJSON), 200) + logger.InfoCF("agent", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), + map[string]any{ + "agent_id": agent.ID, + "tool": tc.Name, + "iteration": iteration, + }) + + // Create async callback for tools that implement AsyncExecutor + asyncCallback := func(callbackCtx context.Context, result *tools.ToolResult) { + if !result.Silent && result.ForUser != "" { + logger.InfoCF("agent", "Async tool completed, agent will handle notification", + map[string]any{ + "tool": tc.Name, + "content_len": len(result.ForUser), + }) + } } - } - toolResult := agent.Tools.ExecuteWithContext( - ctx, - tc.Name, - tc.Arguments, - opts.Channel, - opts.ChatID, - asyncCallback, - ) + toolResult := agent.Tools.ExecuteWithContext( + ctx, + tc.Name, + tc.Arguments, + opts.Channel, + opts.ChatID, + asyncCallback, + ) + agentResults[idx].result = toolResult + }(i, tc) + } + wg.Wait() + // Process results in original order (send to user, save to session) + for _, r := range agentResults { // Send ForUser content to user immediately if not Silent - if !toolResult.Silent && toolResult.ForUser != "" && opts.SendResponse { + if !r.result.Silent && r.result.ForUser != "" && opts.SendResponse { al.bus.PublishOutbound(ctx, bus.OutboundMessage{ Channel: opts.Channel, ChatID: opts.ChatID, - Content: toolResult.ForUser, + Content: r.result.ForUser, }) logger.DebugCF("agent", "Sent tool result to user", map[string]any{ - "tool": tc.Name, - "content_len": len(toolResult.ForUser), + "tool": r.tc.Name, + "content_len": len(r.result.ForUser), }) } // If tool returned media refs, publish them as outbound media - if len(toolResult.Media) > 0 && opts.SendResponse { - parts := make([]bus.MediaPart, 0, len(toolResult.Media)) - for _, ref := range toolResult.Media { + if len(r.result.Media) > 0 && opts.SendResponse { + parts := make([]bus.MediaPart, 0, len(r.result.Media)) + for _, ref := range r.result.Media { part := bus.MediaPart{Ref: ref} - // Populate metadata from MediaStore when available if al.mediaStore != nil { if _, meta, err := al.mediaStore.ResolveWithMeta(ref); err == nil { part.Filename = meta.Filename @@ -1037,15 +1145,15 @@ func (al *AgentLoop) runLLMIteration( } // Determine content for LLM based on tool result - contentForLLM := toolResult.ForLLM - if contentForLLM == "" && toolResult.Err != nil { - contentForLLM = toolResult.Err.Error() + contentForLLM := r.result.ForLLM + if contentForLLM == "" && r.result.Err != nil { + contentForLLM = r.result.Err.Error() } toolResultMsg := providers.Message{ Role: "tool", Content: contentForLLM, - ToolCallID: tc.ID, + ToolCallID: r.tc.ID, } messages = append(messages, toolResultMsg) @@ -1057,33 +1165,13 @@ func (al *AgentLoop) runLLMIteration( return finalContent, iteration, nil } -// updateToolContexts updates the context for tools that need channel/chatID info. -func (al *AgentLoop) updateToolContexts(agent *AgentInstance, channel, chatID string) { - // Use ContextualTool interface instead of type assertions - if tool, ok := agent.Tools.Get("message"); ok { - if mt, ok := tool.(tools.ContextualTool); ok { - mt.SetContext(channel, chatID) - } - } - if tool, ok := agent.Tools.Get("spawn"); ok { - if st, ok := tool.(tools.ContextualTool); ok { - st.SetContext(channel, chatID) - } - } - if tool, ok := agent.Tools.Get("subagent"); ok { - if st, ok := tool.(tools.ContextualTool); ok { - st.SetContext(channel, chatID) - } - } -} - // maybeSummarize triggers summarization if the session history exceeds thresholds. func (al *AgentLoop) maybeSummarize(agent *AgentInstance, sessionKey, channel, chatID string) { newHistory := agent.Sessions.GetHistory(sessionKey) tokenEstimate := al.estimateTokens(newHistory) - threshold := agent.ContextWindow * 75 / 100 + threshold := agent.ContextWindow * agent.SummarizeTokenPercent / 100 - if len(newHistory) > 20 || tokenEstimate > threshold { + if len(newHistory) > agent.SummarizeMessageThreshold || tokenEstimate > threshold { summarizeKey := agent.ID + ":" + sessionKey if _, loading := al.summarizing.LoadOrStore(summarizeKey, true); !loading { go func() { diff --git a/pkg/agent/loop_test.go b/pkg/agent/loop_test.go index 023286f02..aa7d59b5a 100644 --- a/pkg/agent/loop_test.go +++ b/pkg/agent/loop_test.go @@ -164,35 +164,21 @@ func TestToolRegistry_ToolRegistration(t *testing.T) { } } -// TestToolContext_Updates verifies tool context is updated with channel/chatID +// TestToolContext_Updates verifies tool context helpers work correctly func TestToolContext_Updates(t *testing.T) { - tmpDir, err := os.MkdirTemp("", "agent-test-*") - if err != nil { - t.Fatalf("Failed to create temp dir: %v", err) - } - defer os.RemoveAll(tmpDir) + ctx := tools.WithToolContext(context.Background(), "telegram", "chat-42") - cfg := &config.Config{ - Agents: config.AgentsConfig{ - Defaults: config.AgentDefaults{ - Workspace: tmpDir, - Model: "test-model", - MaxTokens: 4096, - MaxToolIterations: 10, - }, - }, + if got := tools.ToolChannel(ctx); got != "telegram" { + t.Errorf("expected channel 'telegram', got %q", got) + } + if got := tools.ToolChatID(ctx); got != "chat-42" { + t.Errorf("expected chatID 'chat-42', got %q", got) } - msgBus := bus.NewMessageBus() - provider := &simpleMockProvider{response: "OK"} - _ = NewAgentLoop(cfg, msgBus, provider) - - // Verify that ContextualTool interface is defined and can be implemented - // This test validates the interface contract exists - ctxTool := &mockContextualTool{} - - // Verify the tool implements the interface correctly - var _ tools.ContextualTool = ctxTool + // Empty context returns empty strings + if got := tools.ToolChannel(context.Background()); got != "" { + t.Errorf("expected empty channel from bare context, got %q", got) + } } // TestToolRegistry_GetDefinitions verifies tool definitions can be retrieved @@ -241,16 +227,11 @@ func TestAgentLoop_GetStartupInfo(t *testing.T) { } defer os.RemoveAll(tmpDir) - cfg := &config.Config{ - Agents: config.AgentsConfig{ - Defaults: config.AgentDefaults{ - Workspace: tmpDir, - Model: "test-model", - MaxTokens: 4096, - MaxToolIterations: 10, - }, - }, - } + cfg := config.DefaultConfig() + cfg.Agents.Defaults.Workspace = tmpDir + cfg.Agents.Defaults.Model = "test-model" + cfg.Agents.Defaults.MaxTokens = 4096 + cfg.Agents.Defaults.MaxToolIterations = 10 msgBus := bus.NewMessageBus() provider := &mockProvider{} @@ -359,36 +340,6 @@ func (m *mockCustomTool) Execute(ctx context.Context, args map[string]any) *tool return tools.SilentResult("Custom tool executed") } -// mockContextualTool tracks context updates -type mockContextualTool struct { - lastChannel string - lastChatID string -} - -func (m *mockContextualTool) Name() string { - return "mock_contextual" -} - -func (m *mockContextualTool) Description() string { - return "Mock contextual tool" -} - -func (m *mockContextualTool) Parameters() map[string]any { - return map[string]any{ - "type": "object", - "properties": map[string]any{}, - } -} - -func (m *mockContextualTool) Execute(ctx context.Context, args map[string]any) *tools.ToolResult { - return tools.SilentResult("Contextual tool executed") -} - -func (m *mockContextualTool) SetContext(channel, chatID string) { - m.lastChannel = channel - m.lastChatID = chatID -} - // testHelper executes a message and returns the response type testHelper struct { al *AgentLoop diff --git a/pkg/agent/thinking.go b/pkg/agent/thinking.go new file mode 100644 index 000000000..015b69282 --- /dev/null +++ b/pkg/agent/thinking.go @@ -0,0 +1,39 @@ +package agent + +import "strings" + +// ThinkingLevel controls how the provider sends thinking parameters. +// +// - "adaptive": sends {thinking: {type: "adaptive"}} + output_config.effort (Claude 4.6+) +// - "low"/"medium"/"high"/"xhigh": sends {thinking: {type: "enabled", budget_tokens: N}} (all models) +// - "off": disables thinking +type ThinkingLevel string + +const ( + ThinkingOff ThinkingLevel = "off" + ThinkingLow ThinkingLevel = "low" + ThinkingMedium ThinkingLevel = "medium" + ThinkingHigh ThinkingLevel = "high" + ThinkingXHigh ThinkingLevel = "xhigh" + ThinkingAdaptive ThinkingLevel = "adaptive" +) + +// parseThinkingLevel normalizes a config string to a ThinkingLevel. +// Case-insensitive and whitespace-tolerant for user-facing config values. +// Returns ThinkingOff for unknown or empty values. +func parseThinkingLevel(level string) ThinkingLevel { + switch strings.ToLower(strings.TrimSpace(level)) { + case "adaptive": + return ThinkingAdaptive + case "low": + return ThinkingLow + case "medium": + return ThinkingMedium + case "high": + return ThinkingHigh + case "xhigh": + return ThinkingXHigh + default: + return ThinkingOff + } +} diff --git a/pkg/agent/thinking_test.go b/pkg/agent/thinking_test.go new file mode 100644 index 000000000..be3a68c33 --- /dev/null +++ b/pkg/agent/thinking_test.go @@ -0,0 +1,35 @@ +package agent + +import "testing" + +func TestParseThinkingLevel(t *testing.T) { + tests := []struct { + name string + input string + want ThinkingLevel + }{ + {"off", "off", ThinkingOff}, + {"empty", "", ThinkingOff}, + {"low", "low", ThinkingLow}, + {"medium", "medium", ThinkingMedium}, + {"high", "high", ThinkingHigh}, + {"xhigh", "xhigh", ThinkingXHigh}, + {"adaptive", "adaptive", ThinkingAdaptive}, + {"unknown", "unknown", ThinkingOff}, + // Case-insensitive and whitespace-tolerant + {"upper_Medium", "Medium", ThinkingMedium}, + {"upper_HIGH", "HIGH", ThinkingHigh}, + {"mixed_Adaptive", "Adaptive", ThinkingAdaptive}, + {"leading_space", " high", ThinkingHigh}, + {"trailing_space", "low ", ThinkingLow}, + {"both_spaces", " medium ", ThinkingMedium}, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + if got := parseThinkingLevel(tt.input); got != tt.want { + t.Errorf("parseThinkingLevel(%q) = %q, want %q", tt.input, got, tt.want) + } + }) + } +} diff --git a/pkg/channels/discord/discord.go b/pkg/channels/discord/discord.go index cd6a2560f..1de910c83 100644 --- a/pkg/channels/discord/discord.go +++ b/pkg/channels/discord/discord.go @@ -3,12 +3,15 @@ package discord import ( "context" "fmt" + "net/http" + "net/url" "os" "strings" "sync" "time" "github.com/bwmarrin/discordgo" + "github.com/gorilla/websocket" "github.com/sipeed/picoclaw/pkg/bus" "github.com/sipeed/picoclaw/pkg/channels" @@ -40,6 +43,9 @@ func NewDiscordChannel(cfg config.DiscordConfig, bus *bus.MessageBus) (*DiscordC return nil, fmt.Errorf("failed to create discord session: %w", err) } + if err := applyDiscordProxy(session, cfg.Proxy); err != nil { + return nil, err + } base := channels.NewBaseChannel("discord", cfg, bus, cfg.AllowFrom, channels.WithMaxMessageLength(2000), channels.WithGroupTrigger(cfg.GroupTrigger), @@ -465,9 +471,43 @@ func (c *DiscordChannel) StartTyping(ctx context.Context, chatID string) (func() func (c *DiscordChannel) downloadAttachment(url, filename string) string { return utils.DownloadFile(url, filename, utils.DownloadOptions{ LoggerPrefix: "discord", + ProxyURL: c.config.Proxy, }) } +func applyDiscordProxy(session *discordgo.Session, proxyAddr string) error { + var proxyFunc func(*http.Request) (*url.URL, error) + if proxyAddr != "" { + proxyURL, err := url.Parse(proxyAddr) + if err != nil { + return fmt.Errorf("invalid discord proxy URL %q: %w", proxyAddr, err) + } + proxyFunc = http.ProxyURL(proxyURL) + } else if os.Getenv("HTTP_PROXY") != "" || os.Getenv("HTTPS_PROXY") != "" { + proxyFunc = http.ProxyFromEnvironment + } + + if proxyFunc == nil { + return nil + } + + transport := &http.Transport{Proxy: proxyFunc} + session.Client = &http.Client{ + Timeout: sendTimeout, + Transport: transport, + } + + if session.Dialer != nil { + dialerCopy := *session.Dialer + dialerCopy.Proxy = proxyFunc + session.Dialer = &dialerCopy + } else { + session.Dialer = &websocket.Dialer{Proxy: proxyFunc} + } + + return nil +} + // stripBotMention removes the bot mention from the message content. // Discord mentions have the format <@USER_ID> or <@!USER_ID> (with nickname). func (c *DiscordChannel) stripBotMention(text string) string { diff --git a/pkg/channels/discord/discord_test.go b/pkg/channels/discord/discord_test.go new file mode 100644 index 000000000..0cd5328f4 --- /dev/null +++ b/pkg/channels/discord/discord_test.go @@ -0,0 +1,91 @@ +package discord + +import ( + "net/http" + "net/url" + "testing" + + "github.com/bwmarrin/discordgo" +) + +func TestApplyDiscordProxy_CustomProxy(t *testing.T) { + session, err := discordgo.New("Bot test-token") + if err != nil { + t.Fatalf("discordgo.New() error: %v", err) + } + + if err = applyDiscordProxy(session, "http://127.0.0.1:7890"); err != nil { + t.Fatalf("applyDiscordProxy() error: %v", err) + } + + req, err := http.NewRequest("GET", "https://discord.com/api/v10/gateway", nil) + if err != nil { + t.Fatalf("http.NewRequest() error: %v", err) + } + + restProxy := session.Client.Transport.(*http.Transport).Proxy + restProxyURL, err := restProxy(req) + if err != nil { + t.Fatalf("rest proxy func error: %v", err) + } + if got, want := restProxyURL.String(), "http://127.0.0.1:7890"; got != want { + t.Fatalf("REST proxy = %q, want %q", got, want) + } + + wsProxyURL, err := session.Dialer.Proxy(req) + if err != nil { + t.Fatalf("ws proxy func error: %v", err) + } + if got, want := wsProxyURL.String(), "http://127.0.0.1:7890"; got != want { + t.Fatalf("WS proxy = %q, want %q", got, want) + } +} + +func TestApplyDiscordProxy_FromEnvironment(t *testing.T) { + t.Setenv("HTTP_PROXY", "http://127.0.0.1:8888") + t.Setenv("http_proxy", "http://127.0.0.1:8888") + t.Setenv("HTTPS_PROXY", "http://127.0.0.1:8888") + t.Setenv("https_proxy", "http://127.0.0.1:8888") + t.Setenv("ALL_PROXY", "") + t.Setenv("all_proxy", "") + t.Setenv("NO_PROXY", "") + t.Setenv("no_proxy", "") + + session, err := discordgo.New("Bot test-token") + if err != nil { + t.Fatalf("discordgo.New() error: %v", err) + } + + if err = applyDiscordProxy(session, ""); err != nil { + t.Fatalf("applyDiscordProxy() error: %v", err) + } + + req, err := http.NewRequest("GET", "https://discord.com/api/v10/gateway", nil) + if err != nil { + t.Fatalf("http.NewRequest() error: %v", err) + } + + gotURL, err := session.Dialer.Proxy(req) + if err != nil { + t.Fatalf("ws proxy func error: %v", err) + } + + wantURL, err := url.Parse("http://127.0.0.1:8888") + if err != nil { + t.Fatalf("url.Parse() error: %v", err) + } + if gotURL.String() != wantURL.String() { + t.Fatalf("WS proxy = %q, want %q", gotURL.String(), wantURL.String()) + } +} + +func TestApplyDiscordProxy_InvalidProxyURL(t *testing.T) { + session, err := discordgo.New("Bot test-token") + if err != nil { + t.Fatalf("discordgo.New() error: %v", err) + } + + if err = applyDiscordProxy(session, "://bad-proxy"); err == nil { + t.Fatal("applyDiscordProxy() expected error for invalid proxy URL, got nil") + } +} diff --git a/pkg/config/config.go b/pkg/config/config.go index 793156be0..259753ab4 100644 --- a/pkg/config/config.go +++ b/pkg/config/config.go @@ -189,6 +189,8 @@ type AgentDefaults struct { MaxTokens int `json:"max_tokens" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_TOKENS"` Temperature *float64 `json:"temperature,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_TEMPERATURE"` MaxToolIterations int `json:"max_tool_iterations" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_TOOL_ITERATIONS"` + SummarizeMessageThreshold int `json:"summarize_message_threshold" env:"PICOCLAW_AGENTS_DEFAULTS_SUMMARIZE_MESSAGE_THRESHOLD"` + SummarizeTokenPercent int `json:"summarize_token_percent" env:"PICOCLAW_AGENTS_DEFAULTS_SUMMARIZE_TOKEN_PERCENT"` MaxMediaSize int `json:"max_media_size,omitempty" env:"PICOCLAW_AGENTS_DEFAULTS_MAX_MEDIA_SIZE"` } @@ -280,6 +282,7 @@ type FeishuConfig struct { type DiscordConfig struct { Enabled bool `json:"enabled" env:"PICOCLAW_CHANNELS_DISCORD_ENABLED"` Token string `json:"token" env:"PICOCLAW_CHANNELS_DISCORD_TOKEN"` + Proxy string `json:"proxy" env:"PICOCLAW_CHANNELS_DISCORD_PROXY"` AllowFrom FlexibleStringSlice `json:"allow_from" env:"PICOCLAW_CHANNELS_DISCORD_ALLOW_FROM"` MentionOnly bool `json:"mention_only" env:"PICOCLAW_CHANNELS_DISCORD_MENTION_ONLY"` GroupTrigger GroupTriggerConfig `json:"group_trigger,omitempty"` @@ -437,6 +440,7 @@ type ProvidersConfig struct { Antigravity ProviderConfig `json:"antigravity"` Qwen ProviderConfig `json:"qwen"` Mistral ProviderConfig `json:"mistral"` + Avian ProviderConfig `json:"avian"` } // IsEmpty checks if all provider configs are empty (no API keys or API bases set) @@ -460,7 +464,8 @@ func (p ProvidersConfig) IsEmpty() bool { p.GitHubCopilot.APIKey == "" && p.GitHubCopilot.APIBase == "" && p.Antigravity.APIKey == "" && p.Antigravity.APIBase == "" && p.Qwen.APIKey == "" && p.Qwen.APIBase == "" && - p.Mistral.APIKey == "" && p.Mistral.APIBase == "" + p.Mistral.APIKey == "" && p.Mistral.APIBase == "" && + p.Avian.APIKey == "" && p.Avian.APIBase == "" } // MarshalJSON implements custom JSON marshaling for ProvidersConfig @@ -511,6 +516,7 @@ type ModelConfig struct { RPM int `json:"rpm,omitempty"` // Requests per minute limit MaxTokensField string `json:"max_tokens_field,omitempty"` // Field name for max tokens (e.g., "max_completion_tokens") RequestTimeout int `json:"request_timeout,omitempty"` + ThinkingLevel string `json:"thinking_level,omitempty"` // Extended thinking: off|low|medium|high|xhigh|adaptive } // Validate checks if the ModelConfig has all required fields. @@ -529,6 +535,10 @@ type GatewayConfig struct { Port int `json:"port" env:"PICOCLAW_GATEWAY_PORT"` } +type ToolConfig struct { + Enabled bool `json:"enabled" env:"ENABLED"` +} + type BraveConfig struct { Enabled bool `json:"enabled" env:"PICOCLAW_TOOLS_WEB_BRAVE_ENABLED"` APIKey string `json:"api_key" env:"PICOCLAW_TOOLS_WEB_BRAVE_API_KEY"` @@ -553,11 +563,23 @@ type PerplexityConfig struct { MaxResults int `json:"max_results" env:"PICOCLAW_TOOLS_WEB_PERPLEXITY_MAX_RESULTS"` } +type GLMSearchConfig struct { + Enabled bool `json:"enabled" env:"PICOCLAW_TOOLS_WEB_GLM_ENABLED"` + APIKey string `json:"api_key" env:"PICOCLAW_TOOLS_WEB_GLM_API_KEY"` + BaseURL string `json:"base_url" env:"PICOCLAW_TOOLS_WEB_GLM_BASE_URL"` + // SearchEngine specifies the search backend: "search_std" (default), + // "search_pro", "search_pro_sogou", or "search_pro_quark". + SearchEngine string `json:"search_engine" env:"PICOCLAW_TOOLS_WEB_GLM_SEARCH_ENGINE"` + MaxResults int `json:"max_results" env:"PICOCLAW_TOOLS_WEB_GLM_MAX_RESULTS"` +} + type WebToolsConfig struct { - Brave BraveConfig `json:"brave"` - Tavily TavilyConfig `json:"tavily"` - DuckDuckGo DuckDuckGoConfig `json:"duckduckgo"` - Perplexity PerplexityConfig `json:"perplexity"` + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_WEB_"` + Brave BraveConfig ` json:"brave"` + Tavily TavilyConfig ` json:"tavily"` + DuckDuckGo DuckDuckGoConfig ` json:"duckduckgo"` + Perplexity PerplexityConfig ` json:"perplexity"` + GLMSearch GLMSearchConfig ` json:"glm_search"` // Proxy is an optional proxy URL for web tools (http/https/socks5/socks5h). // For authenticated proxies, prefer HTTP_PROXY/HTTPS_PROXY env vars instead of embedding credentials in config. Proxy string `json:"proxy,omitempty" env:"PICOCLAW_TOOLS_WEB_PROXY"` @@ -565,19 +587,28 @@ type WebToolsConfig struct { } type CronToolsConfig struct { - ExecTimeoutMinutes int `json:"exec_timeout_minutes" env:"PICOCLAW_TOOLS_CRON_EXEC_TIMEOUT_MINUTES"` // 0 means no timeout + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_CRON_"` + ExecTimeoutMinutes int ` env:"PICOCLAW_TOOLS_CRON_EXEC_TIMEOUT_MINUTES" json:"exec_timeout_minutes"` // 0 means no timeout } type ExecConfig struct { - EnableDenyPatterns bool `json:"enable_deny_patterns" env:"PICOCLAW_TOOLS_EXEC_ENABLE_DENY_PATTERNS"` - CustomDenyPatterns []string `json:"custom_deny_patterns" env:"PICOCLAW_TOOLS_EXEC_CUSTOM_DENY_PATTERNS"` - CustomAllowPatterns []string `json:"custom_allow_patterns" env:"PICOCLAW_TOOLS_EXEC_CUSTOM_ALLOW_PATTERNS"` + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_EXEC_"` + EnableDenyPatterns bool ` env:"PICOCLAW_TOOLS_EXEC_ENABLE_DENY_PATTERNS" json:"enable_deny_patterns"` + CustomDenyPatterns []string ` env:"PICOCLAW_TOOLS_EXEC_CUSTOM_DENY_PATTERNS" json:"custom_deny_patterns"` + CustomAllowPatterns []string ` env:"PICOCLAW_TOOLS_EXEC_CUSTOM_ALLOW_PATTERNS" json:"custom_allow_patterns"` +} + +type SkillsToolsConfig struct { + ToolConfig ` envPrefix:"PICOCLAW_TOOLS_SKILLS_"` + Registries SkillsRegistriesConfig ` json:"registries"` + MaxConcurrentSearches int ` json:"max_concurrent_searches" env:"PICOCLAW_TOOLS_SKILLS_MAX_CONCURRENT_SEARCHES"` + SearchCache SearchCacheConfig ` json:"search_cache"` } type MediaCleanupConfig struct { - Enabled bool `json:"enabled" env:"PICOCLAW_MEDIA_CLEANUP_ENABLED"` - MaxAge int `json:"max_age_minutes" env:"PICOCLAW_MEDIA_CLEANUP_MAX_AGE"` - Interval int `json:"interval_minutes" env:"PICOCLAW_MEDIA_CLEANUP_INTERVAL"` + ToolConfig ` envPrefix:"PICOCLAW_MEDIA_CLEANUP_"` + MaxAge int ` env:"PICOCLAW_MEDIA_CLEANUP_MAX_AGE" json:"max_age_minutes"` + Interval int ` env:"PICOCLAW_MEDIA_CLEANUP_INTERVAL" json:"interval_minutes"` } type ToolsConfig struct { @@ -589,12 +620,19 @@ type ToolsConfig struct { Skills SkillsToolsConfig `json:"skills"` MediaCleanup MediaCleanupConfig `json:"media_cleanup"` MCP MCPConfig `json:"mcp"` -} - -type SkillsToolsConfig struct { - Registries SkillsRegistriesConfig `json:"registries"` - MaxConcurrentSearches int `json:"max_concurrent_searches" env:"PICOCLAW_SKILLS_MAX_CONCURRENT_SEARCHES"` - SearchCache SearchCacheConfig `json:"search_cache"` + AppendFile ToolConfig `json:"append_file" envPrefix:"PICOCLAW_TOOLS_APPEND_FILE_"` + EditFile ToolConfig `json:"edit_file" envPrefix:"PICOCLAW_TOOLS_EDIT_FILE_"` + FindSkills ToolConfig `json:"find_skills" envPrefix:"PICOCLAW_TOOLS_FIND_SKILLS_"` + I2C ToolConfig `json:"i2c" envPrefix:"PICOCLAW_TOOLS_I2C_"` + InstallSkill ToolConfig `json:"install_skill" envPrefix:"PICOCLAW_TOOLS_INSTALL_SKILL_"` + ListDir ToolConfig `json:"list_dir" envPrefix:"PICOCLAW_TOOLS_LIST_DIR_"` + Message ToolConfig `json:"message" envPrefix:"PICOCLAW_TOOLS_MESSAGE_"` + ReadFile ToolConfig `json:"read_file" envPrefix:"PICOCLAW_TOOLS_READ_FILE_"` + Spawn ToolConfig `json:"spawn" envPrefix:"PICOCLAW_TOOLS_SPAWN_"` + SPI ToolConfig `json:"spi" envPrefix:"PICOCLAW_TOOLS_SPI_"` + Subagent ToolConfig `json:"subagent" envPrefix:"PICOCLAW_TOOLS_SUBAGENT_"` + WebFetch ToolConfig `json:"web_fetch" envPrefix:"PICOCLAW_TOOLS_WEB_FETCH_"` + WriteFile ToolConfig `json:"write_file" envPrefix:"PICOCLAW_TOOLS_WRITE_FILE_"` } type SearchCacheConfig struct { @@ -640,8 +678,7 @@ type MCPServerConfig struct { // MCPConfig defines configuration for all MCP servers type MCPConfig struct { - // Enabled globally enables/disables MCP integration - Enabled bool `json:"enabled" env:"PICOCLAW_TOOLS_MCP_ENABLED"` + ToolConfig `envPrefix:"PICOCLAW_TOOLS_MCP_"` // Servers is a map of server name to server configuration Servers map[string]MCPServerConfig `json:"servers,omitempty"` } @@ -895,3 +932,48 @@ func loadProviderEnvOverrides(cfg *Config) { } } } + +func (t *ToolsConfig) IsToolEnabled(name string) bool { + switch name { + case "web": + return t.Web.Enabled + case "cron": + return t.Cron.Enabled + case "exec": + return t.Exec.Enabled + case "skills": + return t.Skills.Enabled + case "media_cleanup": + return t.MediaCleanup.Enabled + case "append_file": + return t.AppendFile.Enabled + case "edit_file": + return t.EditFile.Enabled + case "find_skills": + return t.FindSkills.Enabled + case "i2c": + return t.I2C.Enabled + case "install_skill": + return t.InstallSkill.Enabled + case "list_dir": + return t.ListDir.Enabled + case "message": + return t.Message.Enabled + case "read_file": + return t.ReadFile.Enabled + case "spawn": + return t.Spawn.Enabled + case "spi": + return t.SPI.Enabled + case "subagent": + return t.Subagent.Enabled + case "web_fetch": + return t.WebFetch.Enabled + case "write_file": + return t.WriteFile.Enabled + case "mcp": + return t.MCP.Enabled + default: + return true + } +} diff --git a/pkg/config/config_test.go b/pkg/config/config_test.go index fb11799d4..32add3471 100644 --- a/pkg/config/config_test.go +++ b/pkg/config/config_test.go @@ -436,6 +436,18 @@ func TestLoadConfig_WebToolsProxy(t *testing.T) { } // TestDefaultConfig_DMScope verifies the default dm_scope value +// TestDefaultConfig_SummarizationThresholds verifies summarization defaults +func TestDefaultConfig_SummarizationThresholds(t *testing.T) { + cfg := DefaultConfig() + + if cfg.Agents.Defaults.SummarizeMessageThreshold != 20 { + t.Errorf("SummarizeMessageThreshold = %d, want 20", cfg.Agents.Defaults.SummarizeMessageThreshold) + } + if cfg.Agents.Defaults.SummarizeTokenPercent != 75 { + t.Errorf("SummarizeTokenPercent = %d, want 75", cfg.Agents.Defaults.SummarizeTokenPercent) + } +} + func TestDefaultConfig_DMScope(t *testing.T) { cfg := DefaultConfig() diff --git a/pkg/config/defaults.go b/pkg/config/defaults.go index 9fc09c5f1..488590e28 100644 --- a/pkg/config/defaults.go +++ b/pkg/config/defaults.go @@ -26,13 +26,15 @@ func DefaultConfig() *Config { return &Config{ Agents: AgentsConfig{ Defaults: AgentDefaults{ - Workspace: workspacePath, - RestrictToWorkspace: true, - Provider: "", - Model: "", - MaxTokens: 32768, - Temperature: nil, // nil means use provider default - MaxToolIterations: 50, + Workspace: workspacePath, + RestrictToWorkspace: true, + Provider: "", + Model: "", + MaxTokens: 32768, + Temperature: nil, // nil means use provider default + MaxToolIterations: 50, + SummarizeMessageThreshold: 20, + SummarizeTokenPercent: 75, }, }, Bindings: []AgentBinding{}, @@ -306,6 +308,20 @@ func DefaultConfig() *Config { APIKey: "", }, + // Avian - https://avian.io + { + ModelName: "deepseek-v3.2", + Model: "avian/deepseek/deepseek-v3.2", + APIBase: "https://api.avian.io/v1", + APIKey: "", + }, + { + ModelName: "kimi-k2.5", + Model: "avian/moonshotai/kimi-k2.5", + APIBase: "https://api.avian.io/v1", + APIKey: "", + }, + // VLLM (local) - http://localhost:8000 { ModelName: "local-model", @@ -320,11 +336,16 @@ func DefaultConfig() *Config { }, Tools: ToolsConfig{ MediaCleanup: MediaCleanupConfig{ - Enabled: true, + ToolConfig: ToolConfig{ + Enabled: true, + }, MaxAge: 30, Interval: 5, }, Web: WebToolsConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, Proxy: "", FetchLimitBytes: 10 * 1024 * 1024, // 10MB by default Brave: BraveConfig{ @@ -341,14 +362,30 @@ func DefaultConfig() *Config { APIKey: "", MaxResults: 5, }, + GLMSearch: GLMSearchConfig{ + Enabled: false, + APIKey: "", + BaseURL: "https://open.bigmodel.cn/api/paas/v4/web_search", + SearchEngine: "search_std", + MaxResults: 5, + }, }, Cron: CronToolsConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, ExecTimeoutMinutes: 5, }, Exec: ExecConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, EnableDenyPatterns: true, }, Skills: SkillsToolsConfig{ + ToolConfig: ToolConfig{ + Enabled: true, + }, Registries: SkillsRegistriesConfig{ ClawHub: ClawHubRegistryConfig{ Enabled: true, @@ -362,9 +399,50 @@ func DefaultConfig() *Config { }, }, MCP: MCPConfig{ - Enabled: false, + ToolConfig: ToolConfig{ + Enabled: false, + }, Servers: map[string]MCPServerConfig{}, }, + AppendFile: ToolConfig{ + Enabled: true, + }, + EditFile: ToolConfig{ + Enabled: true, + }, + FindSkills: ToolConfig{ + Enabled: true, + }, + I2C: ToolConfig{ + Enabled: false, // Hardware tool - Linux only + }, + InstallSkill: ToolConfig{ + Enabled: true, + }, + ListDir: ToolConfig{ + Enabled: true, + }, + Message: ToolConfig{ + Enabled: true, + }, + ReadFile: ToolConfig{ + Enabled: true, + }, + Spawn: ToolConfig{ + Enabled: true, + }, + SPI: ToolConfig{ + Enabled: false, // Hardware tool - Linux only + }, + Subagent: ToolConfig{ + Enabled: true, + }, + WebFetch: ToolConfig{ + Enabled: true, + }, + WriteFile: ToolConfig{ + Enabled: true, + }, }, Heartbeat: HeartbeatConfig{ Enabled: true, diff --git a/pkg/config/migration.go b/pkg/config/migration.go index 772f714fd..4a17dd6c9 100644 --- a/pkg/config/migration.go +++ b/pkg/config/migration.go @@ -373,6 +373,23 @@ func ConvertProvidersToModelList(cfg *Config) []ModelConfig { }, true }, }, + { + providerNames: []string{"avian"}, + protocol: "avian", + buildConfig: func(p ProvidersConfig) (ModelConfig, bool) { + if p.Avian.APIKey == "" && p.Avian.APIBase == "" { + return ModelConfig{}, false + } + return ModelConfig{ + ModelName: "avian", + Model: "avian/deepseek/deepseek-v3.2", + APIKey: p.Avian.APIKey, + APIBase: p.Avian.APIBase, + Proxy: p.Avian.Proxy, + RequestTimeout: p.Avian.RequestTimeout, + }, true + }, + }, } // Process each provider migration diff --git a/pkg/config/migration_test.go b/pkg/config/migration_test.go index e24e9fa1d..67ad73db9 100644 --- a/pkg/config/migration_test.go +++ b/pkg/config/migration_test.go @@ -160,14 +160,15 @@ func TestConvertProvidersToModelList_AllProviders(t *testing.T) { Antigravity: ProviderConfig{AuthMethod: "oauth"}, Qwen: ProviderConfig{APIKey: "key17"}, Mistral: ProviderConfig{APIKey: "key18"}, + Avian: ProviderConfig{APIKey: "key19"}, }, } result := ConvertProvidersToModelList(cfg) - // All 19 providers should be converted - if len(result) != 19 { - t.Errorf("len(result) = %d, want 19", len(result)) + // All 20 providers should be converted + if len(result) != 20 { + t.Errorf("len(result) = %d, want 20", len(result)) } } diff --git a/pkg/mcp/manager_test.go b/pkg/mcp/manager_test.go index 8ce81d09e..f353942ab 100644 --- a/pkg/mcp/manager_test.go +++ b/pkg/mcp/manager_test.go @@ -194,7 +194,9 @@ func TestLoadFromMCPConfig_EmptyWorkspaceWithRelativeEnvFile(t *testing.T) { mgr := NewManager() mcpCfg := config.MCPConfig{ - Enabled: true, + ToolConfig: config.ToolConfig{ + Enabled: true, + }, Servers: map[string]config.MCPServerConfig{ "test-server": { Enabled: true, @@ -228,12 +230,20 @@ func TestNewManager_InitialState(t *testing.T) { func TestLoadFromMCPConfig_DisabledOrEmptyServers(t *testing.T) { mgr := NewManager() - err := mgr.LoadFromMCPConfig(context.Background(), config.MCPConfig{Enabled: false}, "/tmp") + err := mgr.LoadFromMCPConfig( + context.Background(), + config.MCPConfig{ToolConfig: config.ToolConfig{Enabled: false}}, + "/tmp", + ) if err != nil { t.Fatalf("expected nil error when MCP disabled, got: %v", err) } - err = mgr.LoadFromMCPConfig(context.Background(), config.MCPConfig{Enabled: true}, "/tmp") + err = mgr.LoadFromMCPConfig( + context.Background(), + config.MCPConfig{ToolConfig: config.ToolConfig{Enabled: true}}, + "/tmp", + ) if err != nil { t.Fatalf("expected nil error when no servers configured, got: %v", err) } diff --git a/pkg/memory/jsonl.go b/pkg/memory/jsonl.go new file mode 100644 index 000000000..e12e2c5ab --- /dev/null +++ b/pkg/memory/jsonl.go @@ -0,0 +1,460 @@ +package memory + +import ( + "bufio" + "bytes" + "context" + "encoding/json" + "fmt" + "hash/fnv" + "log" + "os" + "path/filepath" + "strings" + "sync" + "time" + + "github.com/sipeed/picoclaw/pkg/fileutil" + "github.com/sipeed/picoclaw/pkg/providers" +) + +const ( + // numLockShards is the fixed number of mutexes used to serialize + // per-session access. Using a sharded array instead of a map keeps + // memory bounded regardless of how many sessions are created over + // the lifetime of the process — important for a long-running daemon. + numLockShards = 64 + + // maxLineSize is the maximum size of a single JSON line in a .jsonl + // file. Tool results (read_file, web search, etc.) can be large, so + // we set a generous limit. The scanner starts at 64 KB and grows + // only as needed up to this cap. + maxLineSize = 10 * 1024 * 1024 // 10 MB +) + +// sessionMeta holds per-session metadata stored in a .meta.json file. +type sessionMeta struct { + Key string `json:"key"` + Summary string `json:"summary"` + Skip int `json:"skip"` + Count int `json:"count"` + CreatedAt time.Time `json:"created_at"` + UpdatedAt time.Time `json:"updated_at"` +} + +// JSONLStore implements Store using append-only JSONL files. +// +// Each session is stored as two files: +// +// {sanitized_key}.jsonl — one JSON-encoded message per line, append-only +// {sanitized_key}.meta.json — session metadata (summary, logical truncation offset) +// +// Messages are never physically deleted from the JSONL file. Instead, +// TruncateHistory records a "skip" offset in the metadata file and +// GetHistory ignores lines before that offset. This keeps all writes +// append-only, which is both fast and crash-safe. +type JSONLStore struct { + dir string + locks [numLockShards]sync.Mutex +} + +// NewJSONLStore creates a new JSONL-backed store rooted at dir. +func NewJSONLStore(dir string) (*JSONLStore, error) { + err := os.MkdirAll(dir, 0o755) + if err != nil { + return nil, fmt.Errorf("memory: create directory: %w", err) + } + return &JSONLStore{dir: dir}, nil +} + +// sessionLock returns a mutex for the given session key. +// Keys are mapped to a fixed pool of shards via FNV hash, so +// memory usage is O(1) regardless of total session count. +func (s *JSONLStore) sessionLock(key string) *sync.Mutex { + h := fnv.New32a() + h.Write([]byte(key)) + return &s.locks[h.Sum32()%numLockShards] +} + +func (s *JSONLStore) jsonlPath(key string) string { + return filepath.Join(s.dir, sanitizeKey(key)+".jsonl") +} + +func (s *JSONLStore) metaPath(key string) string { + return filepath.Join(s.dir, sanitizeKey(key)+".meta.json") +} + +// sanitizeKey converts a session key to a safe filename component. +// Mirrors pkg/session.sanitizeFilename so that migration paths match. +// +// Note: this is a lossy mapping — "telegram:123" and "telegram_123" +// both produce the same filename. This is an intentional tradeoff: +// keys with colons (e.g. from channels) are by far the common case, +// and a bidirectional encoding (like URL-encoding) would complicate +// file listings and debugging. +func sanitizeKey(key string) string { + return strings.ReplaceAll(key, ":", "_") +} + +// readMeta loads the metadata file for a session. +// Returns a zero-value sessionMeta if the file does not exist. +func (s *JSONLStore) readMeta(key string) (sessionMeta, error) { + data, err := os.ReadFile(s.metaPath(key)) + if os.IsNotExist(err) { + return sessionMeta{Key: key}, nil + } + if err != nil { + return sessionMeta{}, fmt.Errorf("memory: read meta: %w", err) + } + var meta sessionMeta + err = json.Unmarshal(data, &meta) + if err != nil { + return sessionMeta{}, fmt.Errorf("memory: decode meta: %w", err) + } + return meta, nil +} + +// writeMeta atomically writes the metadata file using the project's +// standard WriteFileAtomic (temp + fsync + rename). +func (s *JSONLStore) writeMeta(key string, meta sessionMeta) error { + data, err := json.MarshalIndent(meta, "", " ") + if err != nil { + return fmt.Errorf("memory: encode meta: %w", err) + } + return fileutil.WriteFileAtomic(s.metaPath(key), data, 0o644) +} + +// readMessages reads valid JSON lines from a .jsonl file, skipping +// the first `skip` lines without unmarshaling them. This avoids the +// cost of json.Unmarshal on logically truncated messages. +// Malformed trailing lines (e.g. from a crash) are silently skipped. +func readMessages(path string, skip int) ([]providers.Message, error) { + f, err := os.Open(path) + if os.IsNotExist(err) { + return []providers.Message{}, nil + } + if err != nil { + return nil, fmt.Errorf("memory: open jsonl: %w", err) + } + defer f.Close() + + var msgs []providers.Message + scanner := bufio.NewScanner(f) + // Allow large lines for tool results (read_file, web search, etc.). + scanner.Buffer(make([]byte, 0, 64*1024), maxLineSize) + + lineNum := 0 + for scanner.Scan() { + line := scanner.Bytes() + if len(line) == 0 { + continue + } + lineNum++ + if lineNum <= skip { + continue + } + var msg providers.Message + if err := json.Unmarshal(line, &msg); err != nil { + // Corrupt line — likely a partial write from a crash. + // Log so operators know data was skipped, but don't + // fail the entire read; this is the standard JSONL + // recovery pattern. + log.Printf("memory: skipping corrupt line %d in %s: %v", + lineNum, filepath.Base(path), err) + continue + } + msgs = append(msgs, msg) + } + if scanner.Err() != nil { + return nil, fmt.Errorf("memory: scan jsonl: %w", scanner.Err()) + } + + if msgs == nil { + msgs = []providers.Message{} + } + return msgs, nil +} + +// countLines counts the total number of non-empty lines in a .jsonl file. +// Used by TruncateHistory to reconcile a stale meta.Count without +// the overhead of unmarshaling every message. +func countLines(path string) (int, error) { + f, err := os.Open(path) + if os.IsNotExist(err) { + return 0, nil + } + if err != nil { + return 0, fmt.Errorf("memory: open jsonl: %w", err) + } + defer f.Close() + + n := 0 + scanner := bufio.NewScanner(f) + scanner.Buffer(make([]byte, 0, 64*1024), maxLineSize) + for scanner.Scan() { + if len(scanner.Bytes()) > 0 { + n++ + } + } + return n, scanner.Err() +} + +func (s *JSONLStore) AddMessage( + _ context.Context, sessionKey, role, content string, +) error { + return s.addMsg(sessionKey, providers.Message{ + Role: role, + Content: content, + }) +} + +func (s *JSONLStore) AddFullMessage( + _ context.Context, sessionKey string, msg providers.Message, +) error { + return s.addMsg(sessionKey, msg) +} + +// addMsg is the shared implementation for AddMessage and AddFullMessage. +func (s *JSONLStore) addMsg(sessionKey string, msg providers.Message) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + // Append the message as a single JSON line. + line, err := json.Marshal(msg) + if err != nil { + return fmt.Errorf("memory: marshal message: %w", err) + } + line = append(line, '\n') + + f, err := os.OpenFile( + s.jsonlPath(sessionKey), + os.O_CREATE|os.O_WRONLY|os.O_APPEND, + 0o644, + ) + if err != nil { + return fmt.Errorf("memory: open jsonl for append: %w", err) + } + _, writeErr := f.Write(line) + if writeErr != nil { + f.Close() + return fmt.Errorf("memory: append message: %w", writeErr) + } + // Flush to physical storage before closing. This matches the + // durability guarantee of writeMeta and rewriteJSONL (which use + // WriteFileAtomic with fsync). Without Sync, a power loss could + // leave the append in the kernel page cache only — lost on reboot. + if syncErr := f.Sync(); syncErr != nil { + f.Close() + return fmt.Errorf("memory: sync jsonl: %w", syncErr) + } + if closeErr := f.Close(); closeErr != nil { + return fmt.Errorf("memory: close jsonl: %w", closeErr) + } + + // Update metadata. + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + now := time.Now() + if meta.Count == 0 && meta.CreatedAt.IsZero() { + meta.CreatedAt = now + } + meta.Count++ + meta.UpdatedAt = now + + return s.writeMeta(sessionKey, meta) +} + +func (s *JSONLStore) GetHistory( + _ context.Context, sessionKey string, +) ([]providers.Message, error) { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return nil, err + } + + // Pass meta.Skip so readMessages skips those lines without + // unmarshaling them — avoids wasted CPU on truncated messages. + msgs, err := readMessages(s.jsonlPath(sessionKey), meta.Skip) + if err != nil { + return nil, err + } + + return msgs, nil +} + +func (s *JSONLStore) GetSummary( + _ context.Context, sessionKey string, +) (string, error) { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return "", err + } + return meta.Summary, nil +} + +func (s *JSONLStore) SetSummary( + _ context.Context, sessionKey, summary string, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + now := time.Now() + if meta.CreatedAt.IsZero() { + meta.CreatedAt = now + } + meta.Summary = summary + meta.UpdatedAt = now + + return s.writeMeta(sessionKey, meta) +} + +func (s *JSONLStore) TruncateHistory( + _ context.Context, sessionKey string, keepLast int, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + + // Always reconcile meta.Count with the actual line count on disk. + // A crash between the JSONL append and the meta update in addMsg + // leaves meta.Count stale (e.g. file has 101 lines but meta says + // 100). Counting lines is cheap — no unmarshal, just a scan — and + // TruncateHistory is not a hot path, so always re-count. + n, countErr := countLines(s.jsonlPath(sessionKey)) + if countErr != nil { + return countErr + } + meta.Count = n + + if keepLast <= 0 { + meta.Skip = meta.Count + } else { + effective := meta.Count - meta.Skip + if keepLast < effective { + meta.Skip = meta.Count - keepLast + } + } + meta.UpdatedAt = time.Now() + + return s.writeMeta(sessionKey, meta) +} + +func (s *JSONLStore) SetHistory( + _ context.Context, + sessionKey string, + history []providers.Message, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + now := time.Now() + if meta.CreatedAt.IsZero() { + meta.CreatedAt = now + } + meta.Skip = 0 + meta.Count = len(history) + meta.UpdatedAt = now + + // Write meta BEFORE rewriting the JSONL file. If we crash between + // the two writes, meta has Skip=0 and the old file is still intact, + // so GetHistory reads from line 1 — returning "too many" messages + // rather than losing data. The next SetHistory call corrects this. + err = s.writeMeta(sessionKey, meta) + if err != nil { + return err + } + + return s.rewriteJSONL(sessionKey, history) +} + +// Compact physically rewrites the JSONL file, dropping all logically +// skipped lines. This reclaims disk space that accumulates after +// repeated TruncateHistory calls. +// +// It is safe to call at any time; if there is nothing to compact +// (skip == 0) the method returns immediately. +func (s *JSONLStore) Compact( + _ context.Context, sessionKey string, +) error { + l := s.sessionLock(sessionKey) + l.Lock() + defer l.Unlock() + + meta, err := s.readMeta(sessionKey) + if err != nil { + return err + } + if meta.Skip == 0 { + return nil + } + + // Read only the active messages, skipping truncated lines + // without unmarshaling them. + active, err := readMessages(s.jsonlPath(sessionKey), meta.Skip) + if err != nil { + return err + } + + // Write meta BEFORE rewriting the JSONL file. If the process + // crashes between the two writes, meta has Skip=0 and the old + // (uncompacted) file is still intact, so GetHistory reads from + // line 1 — returning previously-truncated messages rather than + // losing data. The next Compact or TruncateHistory corrects this. + meta.Skip = 0 + meta.Count = len(active) + meta.UpdatedAt = time.Now() + + err = s.writeMeta(sessionKey, meta) + if err != nil { + return err + } + + return s.rewriteJSONL(sessionKey, active) +} + +// rewriteJSONL atomically replaces the JSONL file with the given messages +// using the project's standard WriteFileAtomic (temp + fsync + rename). +func (s *JSONLStore) rewriteJSONL( + sessionKey string, msgs []providers.Message, +) error { + var buf bytes.Buffer + for i, msg := range msgs { + line, err := json.Marshal(msg) + if err != nil { + return fmt.Errorf("memory: marshal message %d: %w", i, err) + } + buf.Write(line) + buf.WriteByte('\n') + } + return fileutil.WriteFileAtomic(s.jsonlPath(sessionKey), buf.Bytes(), 0o644) +} + +func (s *JSONLStore) Close() error { + return nil +} diff --git a/pkg/memory/jsonl_test.go b/pkg/memory/jsonl_test.go new file mode 100644 index 000000000..356ff14ff --- /dev/null +++ b/pkg/memory/jsonl_test.go @@ -0,0 +1,835 @@ +package memory + +import ( + "context" + "os" + "path/filepath" + "sync" + "testing" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +func newTestStore(t *testing.T) *JSONLStore { + t.Helper() + store, err := NewJSONLStore(t.TempDir()) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + return store +} + +func TestNewJSONLStore_CreatesDirectory(t *testing.T) { + dir := filepath.Join(t.TempDir(), "nested", "sessions") + store, err := NewJSONLStore(dir) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + + info, err := os.Stat(dir) + if err != nil { + t.Fatalf("Stat: %v", err) + } + if !info.IsDir() { + t.Errorf("expected directory, got file") + } +} + +func TestAddMessage_BasicRoundtrip(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + err := store.AddMessage(ctx, "s1", "user", "hello") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + err = store.AddMessage(ctx, "s1", "assistant", "hi there") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "s1") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2 messages, got %d", len(history)) + } + if history[0].Role != "user" || history[0].Content != "hello" { + t.Errorf("msg[0] = %+v", history[0]) + } + if history[1].Role != "assistant" || history[1].Content != "hi there" { + t.Errorf("msg[1] = %+v", history[1]) + } +} + +func TestAddMessage_AutoCreatesSession(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Adding a message to a non-existent session should work. + err := store.AddMessage(ctx, "new-session", "user", "first message") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "new-session") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1 message, got %d", len(history)) + } +} + +func TestAddFullMessage_WithToolCalls(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + msg := providers.Message{ + Role: "assistant", + Content: "Let me search that.", + ToolCalls: []providers.ToolCall{ + { + ID: "call_abc", + Type: "function", + Function: &providers.FunctionCall{ + Name: "web_search", + Arguments: `{"q":"golang jsonl"}`, + }, + }, + }, + } + + err := store.AddFullMessage(ctx, "tc", msg) + if err != nil { + t.Fatalf("AddFullMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "tc") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + if len(history[0].ToolCalls) != 1 { + t.Fatalf("expected 1 tool call, got %d", len(history[0].ToolCalls)) + } + tc := history[0].ToolCalls[0] + if tc.ID != "call_abc" { + t.Errorf("tool call ID = %q", tc.ID) + } + if tc.Function == nil || tc.Function.Name != "web_search" { + t.Errorf("tool call function = %+v", tc.Function) + } +} + +func TestAddFullMessage_ToolCallID(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + msg := providers.Message{ + Role: "tool", + Content: "search results here", + ToolCallID: "call_abc", + } + + err := store.AddFullMessage(ctx, "tr", msg) + if err != nil { + t.Fatalf("AddFullMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "tr") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + if history[0].ToolCallID != "call_abc" { + t.Errorf("ToolCallID = %q", history[0].ToolCallID) + } +} + +func TestGetHistory_EmptySession(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + history, err := store.GetHistory(ctx, "nonexistent") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if history == nil { + t.Fatal("expected non-nil empty slice") + } + if len(history) != 0 { + t.Errorf("expected 0 messages, got %d", len(history)) + } +} + +func TestGetHistory_Ordering(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 5; i++ { + err := store.AddMessage( + ctx, "order", + "user", + string(rune('a'+i)), + ) + if err != nil { + t.Fatalf("AddMessage(%d): %v", i, err) + } + } + + history, err := store.GetHistory(ctx, "order") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 5 { + t.Fatalf("expected 5, got %d", len(history)) + } + for i := 0; i < 5; i++ { + expected := string(rune('a' + i)) + if history[i].Content != expected { + t.Errorf("msg[%d].Content = %q, want %q", i, history[i].Content, expected) + } + } +} + +func TestSetSummary_GetSummary(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // No summary yet. + summary, err := store.GetSummary(ctx, "s1") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "" { + t.Errorf("expected empty, got %q", summary) + } + + // Set a summary. + err = store.SetSummary(ctx, "s1", "talked about Go") + if err != nil { + t.Fatalf("SetSummary: %v", err) + } + + summary, err = store.GetSummary(ctx, "s1") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "talked about Go" { + t.Errorf("summary = %q", summary) + } + + // Update summary. + err = store.SetSummary(ctx, "s1", "updated summary") + if err != nil { + t.Fatalf("SetSummary: %v", err) + } + + summary, err = store.GetSummary(ctx, "s1") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "updated summary" { + t.Errorf("summary = %q", summary) + } +} + +func TestTruncateHistory_KeepLast(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 10; i++ { + err := store.AddMessage( + ctx, "trunc", + "user", + string(rune('a'+i)), + ) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + err := store.TruncateHistory(ctx, "trunc", 4) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "trunc") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 4 { + t.Fatalf("expected 4, got %d", len(history)) + } + // Should be the last 4: g, h, i, j + if history[0].Content != "g" { + t.Errorf("first kept = %q, want 'g'", history[0].Content) + } + if history[3].Content != "j" { + t.Errorf("last kept = %q, want 'j'", history[3].Content) + } +} + +func TestTruncateHistory_KeepZero(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 5; i++ { + err := store.AddMessage(ctx, "empty", "user", "msg") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + err := store.TruncateHistory(ctx, "empty", 0) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "empty") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 0 { + t.Errorf("expected 0, got %d", len(history)) + } +} + +func TestTruncateHistory_KeepMoreThanExists(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 3; i++ { + err := store.AddMessage(ctx, "few", "user", "msg") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Keep 100, but only 3 exist — should keep all. + err := store.TruncateHistory(ctx, "few", 100) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "few") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 3 { + t.Errorf("expected 3, got %d", len(history)) + } +} + +func TestSetHistory_ReplacesAll(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Add some initial messages. + for i := 0; i < 5; i++ { + err := store.AddMessage(ctx, "replace", "user", "old") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Replace with new history. + newHistory := []providers.Message{ + {Role: "user", Content: "new1"}, + {Role: "assistant", Content: "new2"}, + } + err := store.SetHistory(ctx, "replace", newHistory) + if err != nil { + t.Fatalf("SetHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "replace") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2, got %d", len(history)) + } + if history[0].Content != "new1" || history[1].Content != "new2" { + t.Errorf("history = %+v", history) + } +} + +func TestSetHistory_ResetsSkip(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Add messages and truncate. + for i := 0; i < 10; i++ { + err := store.AddMessage(ctx, "skip-reset", "user", "old") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + err := store.TruncateHistory(ctx, "skip-reset", 3) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + // SetHistory should reset skip to 0. + newHistory := []providers.Message{ + {Role: "user", Content: "fresh"}, + } + err = store.SetHistory(ctx, "skip-reset", newHistory) + if err != nil { + t.Fatalf("SetHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "skip-reset") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + if history[0].Content != "fresh" { + t.Errorf("content = %q", history[0].Content) + } +} + +func TestColonInKey(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + err := store.AddMessage(ctx, "telegram:123", "user", "hi") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + history, err := store.GetHistory(ctx, "telegram:123") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1, got %d", len(history)) + } + + // Verify the file is named with underscore. + jsonlFile := filepath.Join(store.dir, "telegram_123.jsonl") + if _, statErr := os.Stat(jsonlFile); statErr != nil { + t.Errorf("expected file %s to exist: %v", jsonlFile, statErr) + } +} + +func TestCompact_RemovesSkippedMessages(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Write 10 messages, then truncate to keep last 3. + for i := 0; i < 10; i++ { + err := store.AddMessage(ctx, "compact", "user", string(rune('a'+i))) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + err := store.TruncateHistory(ctx, "compact", 3) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + // Before compact: file still has 10 lines. + allOnDisk, err := readMessages(store.jsonlPath("compact"), 0) + if err != nil { + t.Fatalf("readMessages: %v", err) + } + if len(allOnDisk) != 10 { + t.Fatalf("before compact: expected 10 on disk, got %d", len(allOnDisk)) + } + + // Compact. + err = store.Compact(ctx, "compact") + if err != nil { + t.Fatalf("Compact: %v", err) + } + + // After compact: file should have only 3 lines. + allOnDisk, err = readMessages(store.jsonlPath("compact"), 0) + if err != nil { + t.Fatalf("readMessages: %v", err) + } + if len(allOnDisk) != 3 { + t.Fatalf("after compact: expected 3 on disk, got %d", len(allOnDisk)) + } + + // GetHistory should still return the same 3 messages. + history, err := store.GetHistory(ctx, "compact") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 3 { + t.Fatalf("expected 3, got %d", len(history)) + } + if history[0].Content != "h" || history[2].Content != "j" { + t.Errorf("wrong content: %+v", history) + } +} + +func TestCompact_NoOpWhenNoSkip(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 5; i++ { + err := store.AddMessage(ctx, "noop", "user", "msg") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Compact without prior truncation — should be a no-op. + err := store.Compact(ctx, "noop") + if err != nil { + t.Fatalf("Compact: %v", err) + } + + history, err := store.GetHistory(ctx, "noop") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 5 { + t.Errorf("expected 5, got %d", len(history)) + } +} + +func TestCompact_ThenAppend(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 8; i++ { + err := store.AddMessage(ctx, "cap", "user", string(rune('a'+i))) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + err := store.TruncateHistory(ctx, "cap", 2) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + err = store.Compact(ctx, "cap") + if err != nil { + t.Fatalf("Compact: %v", err) + } + + // Append after compaction should work correctly. + err = store.AddMessage(ctx, "cap", "user", "new") + if err != nil { + t.Fatalf("AddMessage after compact: %v", err) + } + + history, err := store.GetHistory(ctx, "cap") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 3 { + t.Fatalf("expected 3, got %d", len(history)) + } + // g, h (kept from truncation), new (appended after compaction). + if history[0].Content != "g" { + t.Errorf("first = %q, want 'g'", history[0].Content) + } + if history[2].Content != "new" { + t.Errorf("last = %q, want 'new'", history[2].Content) + } +} + +func TestTruncateHistory_StaleMetaCount(t *testing.T) { + // Simulates a crash between JSONL append and meta update in addMsg: + // file has N+1 lines but meta.Count is still N. TruncateHistory must + // reconcile with the real line count so that keepLast is accurate. + store := newTestStore(t) + ctx := context.Background() + + // Write 10 messages normally (meta.Count = 10). + for i := 0; i < 10; i++ { + err := store.AddMessage(ctx, "stale", "user", string(rune('a'+i))) + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + // Simulate crash: append a line to JSONL but do NOT update meta. + // This leaves meta.Count = 10 while the file has 11 lines. + jsonlPath := store.jsonlPath("stale") + f, err := os.OpenFile(jsonlPath, os.O_WRONLY|os.O_APPEND, 0o644) + if err != nil { + t.Fatalf("open for append: %v", err) + } + _, err = f.WriteString(`{"role":"user","content":"orphan"}` + "\n") + if err != nil { + t.Fatalf("write orphan: %v", err) + } + f.Close() + + // TruncateHistory(keepLast=4) should keep the last 4 of 11 lines, + // not the last 4 of 10. + err = store.TruncateHistory(ctx, "stale", 4) + if err != nil { + t.Fatalf("TruncateHistory: %v", err) + } + + history, err := store.GetHistory(ctx, "stale") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 4 { + t.Fatalf("expected 4, got %d", len(history)) + } + // Last 4 of [a,b,c,d,e,f,g,h,i,j,orphan] = [h,i,j,orphan] + if history[0].Content != "h" { + t.Errorf("first kept = %q, want 'h'", history[0].Content) + } + if history[3].Content != "orphan" { + t.Errorf("last kept = %q, want 'orphan'", history[3].Content) + } +} + +func TestCrashRecovery_PartialLine(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + // Write a valid message first. + err := store.AddMessage(ctx, "crash", "user", "valid") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + // Simulate a crash by appending a partial JSON line directly. + jsonlPath := store.jsonlPath("crash") + f, err := os.OpenFile(jsonlPath, os.O_WRONLY|os.O_APPEND, 0o644) + if err != nil { + t.Fatalf("open for append: %v", err) + } + _, err = f.WriteString(`{"role":"user","content":"incomple`) + if err != nil { + t.Fatalf("write partial: %v", err) + } + f.Close() + + // GetHistory should return only the valid message. + history, err := store.GetHistory(ctx, "crash") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1 valid message, got %d", len(history)) + } + if history[0].Content != "valid" { + t.Errorf("content = %q", history[0].Content) + } +} + +func TestPersistence_AcrossInstances(t *testing.T) { + dir := t.TempDir() + ctx := context.Background() + + // Write with first instance. + store1, err := NewJSONLStore(dir) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + err = store1.AddMessage(ctx, "persist", "user", "remember me") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + err = store1.SetSummary(ctx, "persist", "a test session") + if err != nil { + t.Fatalf("SetSummary: %v", err) + } + store1.Close() + + // Read with second instance. + store2, err := NewJSONLStore(dir) + if err != nil { + t.Fatalf("NewJSONLStore: %v", err) + } + defer store2.Close() + + history, err := store2.GetHistory(ctx, "persist") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 || history[0].Content != "remember me" { + t.Errorf("history = %+v", history) + } + + summary, err := store2.GetSummary(ctx, "persist") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "a test session" { + t.Errorf("summary = %q", summary) + } +} + +func TestConcurrent_AddAndRead(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + var wg sync.WaitGroup + const goroutines = 10 + const msgsPerGoroutine = 20 + + // Concurrent writes. + for g := 0; g < goroutines; g++ { + wg.Add(1) + go func() { + defer wg.Done() + for i := 0; i < msgsPerGoroutine; i++ { + _ = store.AddMessage(ctx, "concurrent", "user", "msg") + } + }() + } + wg.Wait() + + history, err := store.GetHistory(ctx, "concurrent") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + expected := goroutines * msgsPerGoroutine + if len(history) != expected { + t.Errorf("expected %d messages, got %d", expected, len(history)) + } +} + +func TestConcurrent_SummarizeRace(t *testing.T) { + // Simulates the #704 race: one goroutine adds messages while + // another truncates + sets summary — like summarizeSession(). + store := newTestStore(t) + ctx := context.Background() + + // Seed with some messages. + for i := 0; i < 20; i++ { + err := store.AddMessage(ctx, "race", "user", "seed") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + } + + var wg sync.WaitGroup + + // Writer goroutine (main agent loop). + wg.Add(1) + go func() { + defer wg.Done() + for i := 0; i < 50; i++ { + _ = store.AddMessage(ctx, "race", "user", "new") + } + }() + + // Summarizer goroutine (background task). + wg.Add(1) + go func() { + defer wg.Done() + for i := 0; i < 10; i++ { + _ = store.SetSummary(ctx, "race", "summary") + _ = store.TruncateHistory(ctx, "race", 5) + } + }() + + wg.Wait() + + // Verify the store is still in a consistent state. + _, err := store.GetHistory(ctx, "race") + if err != nil { + t.Fatalf("GetHistory after race: %v", err) + } + _, err = store.GetSummary(ctx, "race") + if err != nil { + t.Fatalf("GetSummary after race: %v", err) + } +} + +func TestMultipleSessions_Isolation(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + err := store.AddMessage(ctx, "s1", "user", "msg for s1") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + err = store.AddMessage(ctx, "s2", "user", "msg for s2") + if err != nil { + t.Fatalf("AddMessage: %v", err) + } + + h1, err := store.GetHistory(ctx, "s1") + if err != nil { + t.Fatalf("GetHistory s1: %v", err) + } + h2, err := store.GetHistory(ctx, "s2") + if err != nil { + t.Fatalf("GetHistory s2: %v", err) + } + + if len(h1) != 1 || h1[0].Content != "msg for s1" { + t.Errorf("s1 history = %+v", h1) + } + if len(h2) != 1 || h2[0].Content != "msg for s2" { + t.Errorf("s2 history = %+v", h2) + } +} + +func BenchmarkAddMessage(b *testing.B) { + dir := b.TempDir() + store, err := NewJSONLStore(dir) + if err != nil { + b.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + ctx := context.Background() + + b.ResetTimer() + for i := 0; i < b.N; i++ { + _ = store.AddMessage(ctx, "bench", "user", "benchmark message content") + } +} + +func BenchmarkGetHistory_100(b *testing.B) { + dir := b.TempDir() + store, err := NewJSONLStore(dir) + if err != nil { + b.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + ctx := context.Background() + + for i := 0; i < 100; i++ { + _ = store.AddMessage(ctx, "bench", "user", "message content") + } + + b.ResetTimer() + for i := 0; i < b.N; i++ { + _, _ = store.GetHistory(ctx, "bench") + } +} + +func BenchmarkGetHistory_1000(b *testing.B) { + dir := b.TempDir() + store, err := NewJSONLStore(dir) + if err != nil { + b.Fatalf("NewJSONLStore: %v", err) + } + defer store.Close() + ctx := context.Background() + + for i := 0; i < 1000; i++ { + _ = store.AddMessage(ctx, "bench", "user", "message content") + } + + b.ResetTimer() + for i := 0; i < b.N; i++ { + _, _ = store.GetHistory(ctx, "bench") + } +} diff --git a/pkg/memory/migration.go b/pkg/memory/migration.go new file mode 100644 index 000000000..c9d5176ab --- /dev/null +++ b/pkg/memory/migration.go @@ -0,0 +1,108 @@ +package memory + +import ( + "context" + "encoding/json" + "fmt" + "log" + "os" + "path/filepath" + "strings" + "time" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +// jsonSession mirrors pkg/session.Session for migration purposes. +type jsonSession struct { + Key string `json:"key"` + Messages []providers.Message `json:"messages"` + Summary string `json:"summary,omitempty"` + Created time.Time `json:"created"` + Updated time.Time `json:"updated"` +} + +// MigrateFromJSON reads legacy sessions/*.json files from sessionsDir, +// writes them into the Store, and renames each migrated file to +// .json.migrated as a backup. Returns the number of sessions migrated. +// +// Files that fail to parse are logged and skipped. Already-migrated +// files (.json.migrated) are ignored, making the function idempotent. +func MigrateFromJSON( + ctx context.Context, sessionsDir string, store Store, +) (int, error) { + entries, err := os.ReadDir(sessionsDir) + if os.IsNotExist(err) { + return 0, nil + } + if err != nil { + return 0, fmt.Errorf("memory: read sessions dir: %w", err) + } + + migrated := 0 + for _, entry := range entries { + if entry.IsDir() { + continue + } + name := entry.Name() + if !strings.HasSuffix(name, ".json") { + continue + } + // Skip already-migrated files. + if strings.HasSuffix(name, ".migrated") { + continue + } + + srcPath := filepath.Join(sessionsDir, name) + + data, readErr := os.ReadFile(srcPath) + if readErr != nil { + log.Printf("memory: migrate: skip %s: %v", name, readErr) + continue + } + + var sess jsonSession + if parseErr := json.Unmarshal(data, &sess); parseErr != nil { + log.Printf("memory: migrate: skip %s: %v", name, parseErr) + continue + } + + // Use the key from the JSON content, not the filename. + // Filenames are sanitized (":" → "_") but keys are not. + key := sess.Key + if key == "" { + key = strings.TrimSuffix(name, ".json") + } + + // Use SetHistory (atomic replace) instead of per-message + // AddFullMessage. This makes migration idempotent: if the + // process crashes after writing messages but before the + // rename below, a retry replaces the partial data cleanly + // instead of duplicating messages. + if setErr := store.SetHistory(ctx, key, sess.Messages); setErr != nil { + return migrated, fmt.Errorf( + "memory: migrate %s: set history: %w", + name, setErr, + ) + } + + if sess.Summary != "" { + if sumErr := store.SetSummary(ctx, key, sess.Summary); sumErr != nil { + return migrated, fmt.Errorf( + "memory: migrate %s: set summary: %w", + name, sumErr, + ) + } + } + + // Rename to .migrated as backup (not delete). + renameErr := os.Rename(srcPath, srcPath+".migrated") + if renameErr != nil { + log.Printf("memory: migrate: rename %s: %v", name, renameErr) + } + + migrated++ + } + + return migrated, nil +} diff --git a/pkg/memory/migration_test.go b/pkg/memory/migration_test.go new file mode 100644 index 000000000..3170758b7 --- /dev/null +++ b/pkg/memory/migration_test.go @@ -0,0 +1,384 @@ +package memory + +import ( + "context" + "encoding/json" + "os" + "path/filepath" + "testing" + "time" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +func writeJSONSession( + t *testing.T, dir string, filename string, sess jsonSession, +) { + t.Helper() + data, err := json.MarshalIndent(sess, "", " ") + if err != nil { + t.Fatalf("marshal session: %v", err) + } + err = os.WriteFile(filepath.Join(dir, filename), data, 0o644) + if err != nil { + t.Fatalf("write session file: %v", err) + } +} + +func TestMigrateFromJSON_Basic(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "test.json", jsonSession{ + Key: "test", + Messages: []providers.Message{ + {Role: "user", Content: "hello"}, + {Role: "assistant", Content: "hi"}, + }, + Summary: "A greeting.", + Created: time.Now(), + Updated: time.Now(), + }) + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1 migrated, got %d", count) + } + + history, err := store.GetHistory(ctx, "test") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2 messages, got %d", len(history)) + } + if history[0].Content != "hello" || history[1].Content != "hi" { + t.Errorf("unexpected messages: %+v", history) + } + + summary, err := store.GetSummary(ctx, "test") + if err != nil { + t.Fatalf("GetSummary: %v", err) + } + if summary != "A greeting." { + t.Errorf("summary = %q", summary) + } +} + +func TestMigrateFromJSON_WithToolCalls(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "tools.json", jsonSession{ + Key: "tools", + Messages: []providers.Message{ + { + Role: "assistant", + Content: "Searching...", + ToolCalls: []providers.ToolCall{ + { + ID: "call_1", + Type: "function", + Function: &providers.FunctionCall{ + Name: "web_search", + Arguments: `{"q":"test"}`, + }, + }, + }, + }, + { + Role: "tool", + Content: "result", + ToolCallID: "call_1", + }, + }, + Created: time.Now(), + Updated: time.Now(), + }) + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1, got %d", count) + } + + history, err := store.GetHistory(ctx, "tools") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 2 { + t.Fatalf("expected 2 messages, got %d", len(history)) + } + if len(history[0].ToolCalls) != 1 { + t.Fatalf("expected 1 tool call, got %d", len(history[0].ToolCalls)) + } + if history[0].ToolCalls[0].Function.Name != "web_search" { + t.Errorf("function = %q", history[0].ToolCalls[0].Function.Name) + } + if history[1].ToolCallID != "call_1" { + t.Errorf("ToolCallID = %q", history[1].ToolCallID) + } +} + +func TestMigrateFromJSON_MultipleFiles(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + for i := 0; i < 3; i++ { + key := string(rune('a' + i)) + writeJSONSession(t, sessionsDir, key+".json", jsonSession{ + Key: key, + Messages: []providers.Message{{Role: "user", Content: "msg " + key}}, + Created: time.Now(), + Updated: time.Now(), + }) + } + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 3 { + t.Errorf("expected 3, got %d", count) + } + + for i := 0; i < 3; i++ { + key := string(rune('a' + i)) + history, histErr := store.GetHistory(ctx, key) + if histErr != nil { + t.Fatalf("GetHistory(%q): %v", key, histErr) + } + if len(history) != 1 { + t.Errorf("session %q: expected 1 msg, got %d", key, len(history)) + } + } +} + +func TestMigrateFromJSON_InvalidJSON(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + // One valid, one invalid. + writeJSONSession(t, sessionsDir, "good.json", jsonSession{ + Key: "good", + Messages: []providers.Message{{Role: "user", Content: "ok"}}, + Created: time.Now(), + Updated: time.Now(), + }) + err := os.WriteFile( + filepath.Join(sessionsDir, "bad.json"), + []byte("{invalid json"), + 0o644, + ) + if err != nil { + t.Fatalf("write bad file: %v", err) + } + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1 (bad file skipped), got %d", count) + } + + history, err := store.GetHistory(ctx, "good") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Errorf("expected 1 message, got %d", len(history)) + } +} + +func TestMigrateFromJSON_RenamesFiles(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "rename.json", jsonSession{ + Key: "rename", + Messages: []providers.Message{{Role: "user", Content: "hi"}}, + Created: time.Now(), + Updated: time.Now(), + }) + + _, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + + // Original .json should not exist. + _, statErr := os.Stat(filepath.Join(sessionsDir, "rename.json")) + if !os.IsNotExist(statErr) { + t.Error("rename.json should have been renamed") + } + // .json.migrated should exist. + _, statErr = os.Stat( + filepath.Join(sessionsDir, "rename.json.migrated"), + ) + if statErr != nil { + t.Errorf("rename.json.migrated should exist: %v", statErr) + } +} + +func TestMigrateFromJSON_Idempotent(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "idem.json", jsonSession{ + Key: "idem", + Messages: []providers.Message{{Role: "user", Content: "once"}}, + Created: time.Now(), + Updated: time.Now(), + }) + + count1, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("first migration: %v", err) + } + if count1 != 1 { + t.Errorf("first run: expected 1, got %d", count1) + } + + // Second run should find only .migrated files, skip them. + count2, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("second migration: %v", err) + } + if count2 != 0 { + t.Errorf("second run: expected 0, got %d", count2) + } + + history, err := store.GetHistory(ctx, "idem") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Errorf("expected 1 message, got %d", len(history)) + } +} + +func TestMigrateFromJSON_ColonInKey(t *testing.T) { + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + // File is named telegram_123 (sanitized), but the key inside is telegram:123. + writeJSONSession(t, sessionsDir, "telegram_123.json", jsonSession{ + Key: "telegram:123", + Messages: []providers.Message{{Role: "user", Content: "from telegram"}}, + Created: time.Now(), + Updated: time.Now(), + }) + + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 1 { + t.Errorf("expected 1, got %d", count) + } + + // Accessible via the original key "telegram:123". + history, err := store.GetHistory(ctx, "telegram:123") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history) != 1 { + t.Fatalf("expected 1 message, got %d", len(history)) + } + if history[0].Content != "from telegram" { + t.Errorf("content = %q", history[0].Content) + } + + // In the file-based store, "telegram:123" and "telegram_123" both + // sanitize to the same filename, so they share storage. This is + // expected — the colon-to-underscore mapping is a one-way function. + history2, err := store.GetHistory(ctx, "telegram_123") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + if len(history2) != 1 { + t.Errorf("expected 1 (same file), got %d", len(history2)) + } +} + +func TestMigrateFromJSON_RetryAfterCrash(t *testing.T) { + // Simulates a crash during migration: first run writes messages + // but doesn't rename the .json file. Second run must replace + // (not duplicate) the messages thanks to SetHistory semantics. + sessionsDir := t.TempDir() + store := newTestStore(t) + ctx := context.Background() + + writeJSONSession(t, sessionsDir, "retry.json", jsonSession{ + Key: "retry", + Messages: []providers.Message{ + {Role: "user", Content: "one"}, + {Role: "assistant", Content: "two"}, + }, + Created: time.Now(), + Updated: time.Now(), + }) + + // First migration succeeds — writes messages and renames file. + count, err := MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("first migration: %v", err) + } + if count != 1 { + t.Fatalf("expected 1, got %d", count) + } + + // Simulate "crash before rename": restore the .json file. + src := filepath.Join(sessionsDir, "retry.json.migrated") + dst := filepath.Join(sessionsDir, "retry.json") + if renameErr := os.Rename(src, dst); renameErr != nil { + t.Fatalf("restore .json: %v", renameErr) + } + + // Second migration should re-import without duplicating messages. + count, err = MigrateFromJSON(ctx, sessionsDir, store) + if err != nil { + t.Fatalf("second migration: %v", err) + } + if count != 1 { + t.Fatalf("expected 1, got %d", count) + } + + history, err := store.GetHistory(ctx, "retry") + if err != nil { + t.Fatalf("GetHistory: %v", err) + } + // Must be exactly 2 messages (not 4 from duplication). + if len(history) != 2 { + t.Fatalf("expected 2 messages (no duplicates), got %d", len(history)) + } + if history[0].Content != "one" || history[1].Content != "two" { + t.Errorf("unexpected messages: %+v", history) + } +} + +func TestMigrateFromJSON_NonexistentDir(t *testing.T) { + store := newTestStore(t) + ctx := context.Background() + + count, err := MigrateFromJSON(ctx, "/nonexistent/path", store) + if err != nil { + t.Fatalf("MigrateFromJSON: %v", err) + } + if count != 0 { + t.Errorf("expected 0, got %d", count) + } +} diff --git a/pkg/memory/store.go b/pkg/memory/store.go new file mode 100644 index 000000000..b6e11707d --- /dev/null +++ b/pkg/memory/store.go @@ -0,0 +1,42 @@ +package memory + +import ( + "context" + + "github.com/sipeed/picoclaw/pkg/providers" +) + +// Store defines an interface for persistent session storage. +// Each method is an atomic operation — there is no separate Save() call. +type Store interface { + // AddMessage appends a simple text message to a session. + AddMessage(ctx context.Context, sessionKey, role, content string) error + + // AddFullMessage appends a complete message (with tool calls, etc.) to a session. + AddFullMessage(ctx context.Context, sessionKey string, msg providers.Message) error + + // GetHistory returns all messages for a session in insertion order. + // Returns an empty slice (not nil) if the session does not exist. + GetHistory(ctx context.Context, sessionKey string) ([]providers.Message, error) + + // GetSummary returns the conversation summary for a session. + // Returns an empty string if no summary exists. + GetSummary(ctx context.Context, sessionKey string) (string, error) + + // SetSummary updates the conversation summary for a session. + SetSummary(ctx context.Context, sessionKey, summary string) error + + // TruncateHistory removes all but the last keepLast messages from a session. + // If keepLast <= 0, all messages are removed. + TruncateHistory(ctx context.Context, sessionKey string, keepLast int) error + + // SetHistory replaces all messages in a session with the provided history. + SetHistory(ctx context.Context, sessionKey string, history []providers.Message) error + + // Compact reclaims storage by physically removing logically truncated + // data. Backends that do not accumulate dead data may return nil. + Compact(ctx context.Context, sessionKey string) error + + // Close releases any resources held by the store. + Close() error +} diff --git a/pkg/providers/anthropic/provider.go b/pkg/providers/anthropic/provider.go index 1bb15f771..1b250b9b4 100644 --- a/pkg/providers/anthropic/provider.go +++ b/pkg/providers/anthropic/provider.go @@ -31,6 +31,9 @@ type Provider struct { baseURL string } +// SupportsThinking implements providers.ThinkingCapable. +func (p *Provider) SupportsThinking() bool { return true } + func NewProvider(token string) *Provider { return NewProviderWithBaseURL(token, "") } @@ -182,9 +185,80 @@ func buildParams( params.Tools = translateTools(tools) } + // Extended Thinking / Adaptive Thinking + // The thinking_level value directly determines the API parameter format: + // "adaptive" → {thinking: {type: "adaptive"}} + output_config.effort + // "low/medium/high/xhigh" → {thinking: {type: "enabled", budget_tokens: N}} + if level, ok := options["thinking_level"].(string); ok && level != "" && level != "off" { + applyThinkingConfig(¶ms, level) + } + return params, nil } +// applyThinkingConfig sets thinking parameters based on the level value. +// "adaptive" uses the adaptive thinking API (Claude 4.6+). +// All other levels use budget_tokens which is universally supported. +// +// Anthropic API constraint: temperature must not be set when thinking is enabled. +// budget_tokens must be strictly less than max_tokens. +func applyThinkingConfig(params *anthropic.MessageNewParams, level string) { + // Anthropic API rejects requests with temperature set alongside thinking. + // Reset to zero value (omitted from JSON serialization). + if params.Temperature.Valid() { + log.Printf("anthropic: temperature cleared because thinking is enabled (level=%s)", level) + } + params.Temperature = anthropic.MessageNewParams{}.Temperature + + if level == "adaptive" { + adaptive := anthropic.NewThinkingConfigAdaptiveParam() + params.Thinking = anthropic.ThinkingConfigParamUnion{OfAdaptive: &adaptive} + params.OutputConfig = anthropic.OutputConfigParam{ + Effort: anthropic.OutputConfigEffortHigh, + } + return + } + + budget := int64(levelToBudget(level)) + if budget <= 0 { + return + } + + // budget_tokens must be < max_tokens; clamp to respect user's max_tokens setting. + if budget >= params.MaxTokens { + log.Printf("anthropic: budget_tokens (%d) clamped to %d (max_tokens-1)", budget, params.MaxTokens-1) + budget = params.MaxTokens - 1 + } else if budget > params.MaxTokens*80/100 { + log.Printf("anthropic: thinking budget (%d) exceeds 80%% of max_tokens (%d), output may be truncated", + budget, params.MaxTokens) + } + params.Thinking = anthropic.ThinkingConfigParamOfEnabled(budget) +} + +// levelToBudget maps a thinking level to budget_tokens. +// Values are based on Anthropic's recommendations and community best practices: +// +// low = 4,096 — simple reasoning, quick debugging (Claude Code "think") +// medium = 16,384 — Anthropic recommended sweet spot for most tasks +// high = 32,000 — complex architecture, deep analysis (diminishing returns above this) +// xhigh = 64,000 — extreme reasoning, research problems, benchmarks +// +// Note: For Claude 4.6+, prefer adaptive thinking over manual budget_tokens. +func levelToBudget(level string) int { + switch level { + case "low": + return 4096 + case "medium": + return 16384 + case "high": + return 32000 + case "xhigh": + return 64000 + default: + return 0 + } +} + func translateTools(tools []ToolDefinition) []anthropic.ToolUnionParam { result := make([]anthropic.ToolUnionParam, 0, len(tools)) for _, t := range tools { @@ -213,10 +287,14 @@ func translateTools(tools []ToolDefinition) []anthropic.ToolUnionParam { func parseResponse(resp *anthropic.Message) *LLMResponse { var content strings.Builder + var reasoning strings.Builder var toolCalls []ToolCall for _, block := range resp.Content { switch block.Type { + case "thinking": + tb := block.AsThinking() + reasoning.WriteString(tb.Thinking) case "text": tb := block.AsText() content.WriteString(tb.Text) @@ -247,6 +325,7 @@ func parseResponse(resp *anthropic.Message) *LLMResponse { return &LLMResponse{ Content: content.String(), + Reasoning: reasoning.String(), ToolCalls: toolCalls, FinishReason: finishReason, Usage: &UsageInfo{ diff --git a/pkg/providers/anthropic/thinking_test.go b/pkg/providers/anthropic/thinking_test.go new file mode 100644 index 000000000..e69a3869e --- /dev/null +++ b/pkg/providers/anthropic/thinking_test.go @@ -0,0 +1,212 @@ +package anthropicprovider + +import ( + "encoding/json" + "testing" + + "github.com/anthropics/anthropic-sdk-go" +) + +func TestApplyThinkingConfig_Adaptive(t *testing.T) { + params := anthropic.MessageNewParams{ + MaxTokens: 16000, + Temperature: anthropic.Float(0.7), + } + applyThinkingConfig(¶ms, "adaptive") + + if params.Thinking.OfAdaptive == nil { + t.Fatal("expected adaptive thinking") + } + if params.Thinking.OfEnabled != nil { + t.Error("should not set enabled thinking in adaptive mode") + } + if params.OutputConfig.Effort != anthropic.OutputConfigEffortHigh { + t.Errorf("effort = %q, want %q", params.OutputConfig.Effort, anthropic.OutputConfigEffortHigh) + } + if params.Temperature.Valid() { + t.Error("temperature should be cleared when thinking is enabled") + } +} + +func TestApplyThinkingConfig_BudgetLevels(t *testing.T) { + tests := []struct { + level string + wantBudget int64 + }{ + {"low", 4096}, + {"medium", 16384}, + {"high", 32000}, + {"xhigh", 64000}, + } + + for _, tt := range tests { + t.Run(tt.level, func(t *testing.T) { + params := anthropic.MessageNewParams{ + MaxTokens: 200000, + Temperature: anthropic.Float(0.5), + } + applyThinkingConfig(¶ms, tt.level) + + if params.Thinking.OfEnabled == nil { + t.Fatal("expected enabled thinking") + } + if params.Thinking.OfAdaptive != nil { + t.Error("should not set adaptive thinking") + } + if params.Thinking.OfEnabled.BudgetTokens != tt.wantBudget { + t.Errorf("budget_tokens = %d, want %d", params.Thinking.OfEnabled.BudgetTokens, tt.wantBudget) + } + if params.OutputConfig.Effort != "" { + t.Errorf("effort = %q, want empty", params.OutputConfig.Effort) + } + if params.Temperature.Valid() { + t.Error("temperature should be cleared when thinking is enabled") + } + }) + } +} + +func TestApplyThinkingConfig_BudgetClamp(t *testing.T) { + // budget_tokens must be < max_tokens; clamp budget down to respect user's max_tokens. + params := anthropic.MessageNewParams{MaxTokens: 4096} + applyThinkingConfig(¶ms, "high") // budget=32000 > maxTokens=4096 + + if params.Thinking.OfEnabled == nil { + t.Fatal("expected enabled thinking") + } + if params.Thinking.OfEnabled.BudgetTokens != 4095 { + t.Errorf("budget_tokens = %d, want 4095 (maxTokens-1)", params.Thinking.OfEnabled.BudgetTokens) + } + if params.MaxTokens != 4096 { + t.Errorf("max_tokens should not be modified, got %d", params.MaxTokens) + } +} + +func TestApplyThinkingConfig_UnknownLevel(t *testing.T) { + params := anthropic.MessageNewParams{MaxTokens: 16000} + applyThinkingConfig(¶ms, "unknown") + + if params.Thinking.OfEnabled != nil { + t.Error("should not set enabled thinking for unknown level") + } + if params.Thinking.OfAdaptive != nil { + t.Error("should not set adaptive thinking for unknown level") + } +} + +func TestLevelToBudget(t *testing.T) { + tests := []struct { + name string + level string + want int + }{ + {"low", "low", 4096}, + {"medium", "medium", 16384}, + {"high", "high", 32000}, + {"xhigh", "xhigh", 64000}, + {"off", "off", 0}, + {"empty", "", 0}, + } + + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + if got := levelToBudget(tt.level); got != tt.want { + t.Errorf("levelToBudget(%q) = %d, want %d", tt.level, got, tt.want) + } + }) + } +} + +func TestBuildParams_ThinkingClearsTemperature(t *testing.T) { + msgs := []Message{{Role: "user", Content: "hello"}} + opts := map[string]any{ + "max_tokens": 200000, + "temperature": 0.8, + "thinking_level": "medium", + } + + params, err := buildParams(msgs, nil, "claude-sonnet-4-6", opts) + if err != nil { + t.Fatal(err) + } + + if params.Temperature.Valid() { + t.Error("temperature should be cleared when thinking_level is set") + } + if params.Thinking.OfEnabled == nil { + t.Fatal("expected enabled thinking") + } + if params.Thinking.OfEnabled.BudgetTokens != 16384 { + t.Errorf("budget_tokens = %d, want 16384", params.Thinking.OfEnabled.BudgetTokens) + } +} + +// unmarshalBlocks constructs []ContentBlockUnion via JSON round-trip so that +// the internal JSON.raw field is populated (required by AsText/AsThinking). +func unmarshalBlocks(t *testing.T, jsonStr string) []anthropic.ContentBlockUnion { + t.Helper() + var blocks []anthropic.ContentBlockUnion + if err := json.Unmarshal([]byte(jsonStr), &blocks); err != nil { + t.Fatalf("unmarshalBlocks: %v", err) + } + return blocks +} + +func TestParseResponse_ThinkingBlock(t *testing.T) { + resp := &anthropic.Message{ + Content: unmarshalBlocks(t, `[ + {"type":"thinking","thinking":"Let me reason step by step...","signature":"sig"}, + {"type":"text","text":"The answer is 42."} + ]`), + StopReason: anthropic.StopReasonEndTurn, + } + + result := parseResponse(resp) + + if result.Reasoning != "Let me reason step by step..." { + t.Errorf("Reasoning = %q, want thinking content", result.Reasoning) + } + if result.Content != "The answer is 42." { + t.Errorf("Content = %q, want text content", result.Content) + } + if result.FinishReason != "stop" { + t.Errorf("FinishReason = %q, want stop", result.FinishReason) + } +} + +func TestParseResponse_NoThinkingBlock(t *testing.T) { + resp := &anthropic.Message{ + Content: unmarshalBlocks(t, `[ + {"type":"text","text":"Just a normal response."} + ]`), + StopReason: anthropic.StopReasonEndTurn, + } + + result := parseResponse(resp) + + if result.Reasoning != "" { + t.Errorf("Reasoning = %q, want empty", result.Reasoning) + } + if result.Content != "Just a normal response." { + t.Errorf("Content = %q, want text content", result.Content) + } +} + +func TestBuildParams_NoThinkingKeepsTemperature(t *testing.T) { + msgs := []Message{{Role: "user", Content: "hello"}} + opts := map[string]any{ + "temperature": 0.8, + } + + params, err := buildParams(msgs, nil, "claude-sonnet-4-6", opts) + if err != nil { + t.Fatal(err) + } + + if !params.Temperature.Valid() { + t.Error("temperature should be preserved when thinking is not set") + } + if params.Temperature.Value != 0.8 { + t.Errorf("temperature = %f, want 0.8", params.Temperature.Value) + } +} diff --git a/pkg/providers/factory.go b/pkg/providers/factory.go index 5b3e42b9e..a0d09a835 100644 --- a/pkg/providers/factory.go +++ b/pkg/providers/factory.go @@ -181,6 +181,15 @@ func resolveProviderSelection(cfg *config.Config) (providerSelection, error) { sel.model = "deepseek-chat" } } + case "avian": + if cfg.Providers.Avian.APIKey != "" { + sel.apiKey = cfg.Providers.Avian.APIKey + sel.apiBase = cfg.Providers.Avian.APIBase + sel.proxy = cfg.Providers.Avian.Proxy + if sel.apiBase == "" { + sel.apiBase = "https://api.avian.io/v1" + } + } case "mistral": if cfg.Providers.Mistral.APIKey != "" { sel.apiKey = cfg.Providers.Mistral.APIKey @@ -300,6 +309,13 @@ func resolveProviderSelection(cfg *config.Config) (providerSelection, error) { if sel.apiBase == "" { sel.apiBase = "https://api.mistral.ai/v1" } + case strings.HasPrefix(model, "avian/") && cfg.Providers.Avian.APIKey != "": + sel.apiKey = cfg.Providers.Avian.APIKey + sel.apiBase = cfg.Providers.Avian.APIBase + sel.proxy = cfg.Providers.Avian.Proxy + if sel.apiBase == "" { + sel.apiBase = "https://api.avian.io/v1" + } case cfg.Providers.VLLM.APIBase != "": sel.apiKey = cfg.Providers.VLLM.APIKey sel.apiBase = cfg.Providers.VLLM.APIBase diff --git a/pkg/providers/factory_provider.go b/pkg/providers/factory_provider.go index 155317a3b..c05fb0ad4 100644 --- a/pkg/providers/factory_provider.go +++ b/pkg/providers/factory_provider.go @@ -94,7 +94,7 @@ func CreateProviderFromConfig(cfg *config.ModelConfig) (LLMProvider, string, err case "litellm", "openrouter", "groq", "zhipu", "gemini", "nvidia", "ollama", "moonshot", "shengsuanyun", "deepseek", "cerebras", - "volcengine", "vllm", "qwen", "mistral": + "volcengine", "vllm", "qwen", "mistral", "avian": // All other OpenAI-compatible HTTP providers if cfg.APIKey == "" && cfg.APIBase == "" { return nil, "", fmt.Errorf("api_key or api_base is required for HTTP-based protocol %q", protocol) @@ -208,6 +208,8 @@ func getDefaultAPIBase(protocol string) string { return "http://localhost:8000/v1" case "mistral": return "https://api.mistral.ai/v1" + case "avian": + return "https://api.avian.io/v1" default: return "" } diff --git a/pkg/providers/openai_compat/provider.go b/pkg/providers/openai_compat/provider.go index ff9109e96..1904ee153 100644 --- a/pkg/providers/openai_compat/provider.go +++ b/pkg/providers/openai_compat/provider.go @@ -323,12 +323,14 @@ func serializeMessages(messages []Message) []any { }) } for _, mediaURL := range m.Media { - parts = append(parts, map[string]any{ - "type": "image_url", - "image_url": map[string]any{ - "url": mediaURL, - }, - }) + if strings.HasPrefix(mediaURL, "data:image/") { + parts = append(parts, map[string]any{ + "type": "image_url", + "image_url": map[string]any{ + "url": mediaURL, + }, + }) + } } msg := map[string]any{ diff --git a/pkg/providers/types.go b/pkg/providers/types.go index f0c168bc6..68bbd1e65 100644 --- a/pkg/providers/types.go +++ b/pkg/providers/types.go @@ -37,6 +37,13 @@ type StatefulProvider interface { Close() } +// ThinkingCapable is an optional interface for providers that support +// extended thinking (e.g. Anthropic). Used by the agent loop to warn +// when thinking_level is configured but the active provider cannot use it. +type ThinkingCapable interface { + SupportsThinking() bool +} + // FailoverReason classifies why an LLM request failed for fallback decisions. type FailoverReason string diff --git a/pkg/tools/base.go b/pkg/tools/base.go index 770d8cb04..ec743e164 100644 --- a/pkg/tools/base.go +++ b/pkg/tools/base.go @@ -10,11 +10,38 @@ type Tool interface { Execute(ctx context.Context, args map[string]any) *ToolResult } -// ContextualTool is an optional interface that tools can implement -// to receive the current message context (channel, chatID) -type ContextualTool interface { - Tool - SetContext(channel, chatID string) +// --- Request-scoped tool context (channel / chatID) --- +// +// Carried via context.Value so that concurrent tool calls each receive +// their own immutable copy — no mutable state on singleton tool instances. +// +// Keys are unexported pointer-typed vars — guaranteed collision-free, +// and only accessible through the helper functions below. + +type toolCtxKey struct{ name string } + +var ( + ctxKeyChannel = &toolCtxKey{"channel"} + ctxKeyChatID = &toolCtxKey{"chatID"} +) + +// WithToolContext returns a child context carrying channel and chatID. +func WithToolContext(ctx context.Context, channel, chatID string) context.Context { + ctx = context.WithValue(ctx, ctxKeyChannel, channel) + ctx = context.WithValue(ctx, ctxKeyChatID, chatID) + return ctx +} + +// ToolChannel extracts the channel from ctx, or "" if unset. +func ToolChannel(ctx context.Context) string { + v, _ := ctx.Value(ctxKeyChannel).(string) + return v +} + +// ToolChatID extracts the chatID from ctx, or "" if unset. +func ToolChatID(ctx context.Context) string { + v, _ := ctx.Value(ctxKeyChatID).(string) + return v } // AsyncCallback is a function type that async tools use to notify completion. @@ -22,51 +49,36 @@ type ContextualTool interface { // // The ctx parameter allows the callback to be canceled if the agent is shutting down. // The result parameter contains the tool's execution result. -// -// Example usage in an async tool: -// -// func (t *MyAsyncTool) Execute(ctx context.Context, args map[string]interface{}) *ToolResult { -// // Start async work in background -// go func() { -// result := doAsyncWork() -// if t.callback != nil { -// t.callback(ctx, result) -// } -// }() -// return AsyncResult("Async task started") -// } type AsyncCallback func(ctx context.Context, result *ToolResult) -// AsyncTool is an optional interface that tools can implement to support +// AsyncExecutor is an optional interface that tools can implement to support // asynchronous execution with completion callbacks. // -// Async tools return immediately with an AsyncResult, then notify completion -// via the callback set by SetCallback. +// Unlike the old AsyncTool pattern (SetCallback + Execute), AsyncExecutor +// receives the callback as a parameter of ExecuteAsync. This eliminates the +// data race where concurrent calls could overwrite each other's callbacks +// on a shared tool instance. // // This is useful for: -// - Long-running operations that shouldn't block the agent loop -// - Subagent spawns that complete independently -// - Background tasks that need to report results later +// - Long-running operations that shouldn't block the agent loop +// - Subagent spawns that complete independently +// - Background tasks that need to report results later // // Example: // -// type SpawnTool struct { -// callback AsyncCallback -// } -// -// func (t *SpawnTool) SetCallback(cb AsyncCallback) { -// t.callback = cb -// } -// -// func (t *SpawnTool) Execute(ctx context.Context, args map[string]interface{}) *ToolResult { -// go t.runSubagent(ctx, args) +// func (t *SpawnTool) ExecuteAsync(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult { +// go func() { +// result := t.runSubagent(ctx, args) +// if cb != nil { cb(ctx, result) } +// }() // return AsyncResult("Subagent spawned, will report back") // } -type AsyncTool interface { +type AsyncExecutor interface { Tool - // SetCallback registers a callback function to be invoked when the async operation completes. - // The callback will be called from a goroutine and should handle thread-safety if needed. - SetCallback(cb AsyncCallback) + // ExecuteAsync runs the tool asynchronously. The callback cb will be + // invoked (possibly from another goroutine) when the async operation + // completes. cb is guaranteed to be non-nil by the caller (registry). + ExecuteAsync(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult } func ToolToSchema(tool Tool) map[string]any { diff --git a/pkg/tools/cron.go b/pkg/tools/cron.go index 6888d1326..31ac9ab88 100644 --- a/pkg/tools/cron.go +++ b/pkg/tools/cron.go @@ -4,7 +4,6 @@ import ( "context" "fmt" "strings" - "sync" "time" "github.com/sipeed/picoclaw/pkg/bus" @@ -24,9 +23,6 @@ type CronTool struct { executor JobExecutor msgBus *bus.MessageBus execTool *ExecTool - channel string - chatID string - mu sync.RWMutex } // NewCronTool creates a new CronTool @@ -102,14 +98,6 @@ func (t *CronTool) Parameters() map[string]any { } } -// SetContext sets the current session context for job creation -func (t *CronTool) SetContext(channel, chatID string) { - t.mu.Lock() - defer t.mu.Unlock() - t.channel = channel - t.chatID = chatID -} - // Execute runs the tool with the given arguments func (t *CronTool) Execute(ctx context.Context, args map[string]any) *ToolResult { action, ok := args["action"].(string) @@ -119,7 +107,7 @@ func (t *CronTool) Execute(ctx context.Context, args map[string]any) *ToolResult switch action { case "add": - return t.addJob(args) + return t.addJob(ctx, args) case "list": return t.listJobs() case "remove": @@ -133,11 +121,9 @@ func (t *CronTool) Execute(ctx context.Context, args map[string]any) *ToolResult } } -func (t *CronTool) addJob(args map[string]any) *ToolResult { - t.mu.RLock() - channel := t.channel - chatID := t.chatID - t.mu.RUnlock() +func (t *CronTool) addJob(ctx context.Context, args map[string]any) *ToolResult { + channel := ToolChannel(ctx) + chatID := ToolChatID(ctx) if channel == "" || chatID == "" { return ErrorResult("no session context (channel/chat_id not set). Use this tool in an active conversation.") diff --git a/pkg/tools/message.go b/pkg/tools/message.go index 15ef4ff73..438ceeddd 100644 --- a/pkg/tools/message.go +++ b/pkg/tools/message.go @@ -3,15 +3,14 @@ package tools import ( "context" "fmt" + "sync/atomic" ) type SendCallback func(channel, chatID, content string) error type MessageTool struct { - sendCallback SendCallback - defaultChannel string - defaultChatID string - sentInRound bool // Tracks whether a message was sent in the current processing round + sendCallback SendCallback + sentInRound atomic.Bool // Tracks whether a message was sent in the current processing round } func NewMessageTool() *MessageTool { @@ -47,15 +46,15 @@ func (t *MessageTool) Parameters() map[string]any { } } -func (t *MessageTool) SetContext(channel, chatID string) { - t.defaultChannel = channel - t.defaultChatID = chatID - t.sentInRound = false // Reset send tracking for new processing round +// ResetSentInRound resets the per-round send tracker. +// Called by the agent loop at the start of each inbound message processing round. +func (t *MessageTool) ResetSentInRound() { + t.sentInRound.Store(false) } // HasSentInRound returns true if the message tool sent a message during the current round. func (t *MessageTool) HasSentInRound() bool { - return t.sentInRound + return t.sentInRound.Load() } func (t *MessageTool) SetSendCallback(callback SendCallback) { @@ -72,10 +71,10 @@ func (t *MessageTool) Execute(ctx context.Context, args map[string]any) *ToolRes chatID, _ := args["chat_id"].(string) if channel == "" { - channel = t.defaultChannel + channel = ToolChannel(ctx) } if chatID == "" { - chatID = t.defaultChatID + chatID = ToolChatID(ctx) } if channel == "" || chatID == "" { @@ -94,7 +93,7 @@ func (t *MessageTool) Execute(ctx context.Context, args map[string]any) *ToolRes } } - t.sentInRound = true + t.sentInRound.Store(true) // Silent: user already received the message directly return &ToolResult{ ForLLM: fmt.Sprintf("Message sent to %s:%s", channel, chatID), diff --git a/pkg/tools/message_test.go b/pkg/tools/message_test.go index 717c1117b..05630972e 100644 --- a/pkg/tools/message_test.go +++ b/pkg/tools/message_test.go @@ -8,7 +8,6 @@ import ( func TestMessageTool_Execute_Success(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") var sentChannel, sentChatID, sentContent string tool.SetSendCallback(func(channel, chatID, content string) error { @@ -18,7 +17,7 @@ func TestMessageTool_Execute_Success(t *testing.T) { return nil }) - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{ "content": "Hello, world!", } @@ -60,7 +59,6 @@ func TestMessageTool_Execute_Success(t *testing.T) { func TestMessageTool_Execute_WithCustomChannel(t *testing.T) { tool := NewMessageTool() - tool.SetContext("default-channel", "default-chat-id") var sentChannel, sentChatID string tool.SetSendCallback(func(channel, chatID, content string) error { @@ -69,7 +67,7 @@ func TestMessageTool_Execute_WithCustomChannel(t *testing.T) { return nil }) - ctx := context.Background() + ctx := WithToolContext(context.Background(), "default-channel", "default-chat-id") args := map[string]any{ "content": "Test message", "channel": "custom-channel", @@ -96,14 +94,13 @@ func TestMessageTool_Execute_WithCustomChannel(t *testing.T) { func TestMessageTool_Execute_SendFailure(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") sendErr := errors.New("network error") tool.SetSendCallback(func(channel, chatID, content string) error { return sendErr }) - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{ "content": "Test message", } @@ -133,9 +130,8 @@ func TestMessageTool_Execute_SendFailure(t *testing.T) { func TestMessageTool_Execute_MissingContent(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{} // content missing result := tool.Execute(ctx, args) @@ -151,7 +147,7 @@ func TestMessageTool_Execute_MissingContent(t *testing.T) { func TestMessageTool_Execute_NoTargetChannel(t *testing.T) { tool := NewMessageTool() - // No SetContext called, so defaultChannel and defaultChatID are empty + // No WithToolContext — channel/chatID are empty tool.SetSendCallback(func(channel, chatID, content string) error { return nil @@ -175,10 +171,9 @@ func TestMessageTool_Execute_NoTargetChannel(t *testing.T) { func TestMessageTool_Execute_NotConfigured(t *testing.T) { tool := NewMessageTool() - tool.SetContext("test-channel", "test-chat-id") // No SetSendCallback called - ctx := context.Background() + ctx := WithToolContext(context.Background(), "test-channel", "test-chat-id") args := map[string]any{ "content": "Test message", } diff --git a/pkg/tools/registry.go b/pkg/tools/registry.go index 0ba983e02..ca8436c67 100644 --- a/pkg/tools/registry.go +++ b/pkg/tools/registry.go @@ -45,8 +45,9 @@ func (r *ToolRegistry) Execute(ctx context.Context, name string, args map[string } // ExecuteWithContext executes a tool with channel/chatID context and optional async callback. -// If the tool implements AsyncTool and a non-nil callback is provided, -// the callback will be set on the tool before execution. +// If the tool implements AsyncExecutor and a non-nil callback is provided, +// ExecuteAsync is called instead of Execute — the callback is a parameter, +// never stored as mutable state on the tool. func (r *ToolRegistry) ExecuteWithContext( ctx context.Context, name string, @@ -69,22 +70,23 @@ func (r *ToolRegistry) ExecuteWithContext( return ErrorResult(fmt.Sprintf("tool %q not found", name)).WithError(fmt.Errorf("tool not found")) } - // If tool implements ContextualTool, set context - if contextualTool, ok := tool.(ContextualTool); ok && channel != "" && chatID != "" { - contextualTool.SetContext(channel, chatID) - } + // Inject channel/chatID into ctx so tools read them via ToolChannel(ctx)/ToolChatID(ctx). + // Always inject — tools validate what they require. + ctx = WithToolContext(ctx, channel, chatID) - // If tool implements AsyncTool and callback is provided, set callback - if asyncTool, ok := tool.(AsyncTool); ok && asyncCallback != nil { - asyncTool.SetCallback(asyncCallback) - logger.DebugCF("tool", "Async callback injected", + // If tool implements AsyncExecutor and callback is provided, use ExecuteAsync. + // The callback is a call parameter, not mutable state on the tool instance. + var result *ToolResult + start := time.Now() + if asyncExec, ok := tool.(AsyncExecutor); ok && asyncCallback != nil { + logger.DebugCF("tool", "Executing async tool via ExecuteAsync", map[string]any{ "tool": name, }) + result = asyncExec.ExecuteAsync(ctx, args, asyncCallback) + } else { + result = tool.Execute(ctx, args) } - - start := time.Now() - result := tool.Execute(ctx, args) duration := time.Since(start) // Log based on result type diff --git a/pkg/tools/registry_test.go b/pkg/tools/registry_test.go index 8fe88ca78..92d7d5abd 100644 --- a/pkg/tools/registry_test.go +++ b/pkg/tools/registry_test.go @@ -25,24 +25,24 @@ func (m *mockRegistryTool) Execute(_ context.Context, _ map[string]any) *ToolRes return m.result } -type mockCtxTool struct { +type mockContextAwareTool struct { mockRegistryTool - channel string - chatID string + lastCtx context.Context } -func (m *mockCtxTool) SetContext(channel, chatID string) { - m.channel = channel - m.chatID = chatID +func (m *mockContextAwareTool) Execute(ctx context.Context, _ map[string]any) *ToolResult { + m.lastCtx = ctx + return m.result } type mockAsyncRegistryTool struct { mockRegistryTool - cb AsyncCallback + lastCB AsyncCallback } -func (m *mockAsyncRegistryTool) SetCallback(cb AsyncCallback) { - m.cb = cb +func (m *mockAsyncRegistryTool) ExecuteAsync(_ context.Context, args map[string]any, cb AsyncCallback) *ToolResult { + m.lastCB = cb + return m.result } // --- helpers --- @@ -136,34 +136,44 @@ func TestToolRegistry_Execute_NotFound(t *testing.T) { } } -func TestToolRegistry_ExecuteWithContext_ContextualTool(t *testing.T) { +func TestToolRegistry_ExecuteWithContext_InjectsToolContext(t *testing.T) { r := NewToolRegistry() - ct := &mockCtxTool{ + ct := &mockContextAwareTool{ mockRegistryTool: *newMockTool("ctx_tool", "needs context"), } r.Register(ct) r.ExecuteWithContext(context.Background(), "ctx_tool", nil, "telegram", "chat-42", nil) - if ct.channel != "telegram" { - t.Errorf("expected channel 'telegram', got %q", ct.channel) + if ct.lastCtx == nil { + t.Fatal("expected Execute to be called") } - if ct.chatID != "chat-42" { - t.Errorf("expected chatID 'chat-42', got %q", ct.chatID) + if got := ToolChannel(ct.lastCtx); got != "telegram" { + t.Errorf("expected channel 'telegram', got %q", got) + } + if got := ToolChatID(ct.lastCtx); got != "chat-42" { + t.Errorf("expected chatID 'chat-42', got %q", got) } } -func TestToolRegistry_ExecuteWithContext_SkipsEmptyContext(t *testing.T) { +func TestToolRegistry_ExecuteWithContext_EmptyContext(t *testing.T) { r := NewToolRegistry() - ct := &mockCtxTool{ + ct := &mockContextAwareTool{ mockRegistryTool: *newMockTool("ctx_tool", "needs context"), } r.Register(ct) r.ExecuteWithContext(context.Background(), "ctx_tool", nil, "", "", nil) - if ct.channel != "" || ct.chatID != "" { - t.Error("SetContext should not be called with empty channel/chatID") + if ct.lastCtx == nil { + t.Fatal("expected Execute to be called") + } + // Empty values are still injected; tools decide what to do with them. + if got := ToolChannel(ct.lastCtx); got != "" { + t.Errorf("expected empty channel, got %q", got) + } + if got := ToolChatID(ct.lastCtx); got != "" { + t.Errorf("expected empty chatID, got %q", got) } } @@ -179,14 +189,14 @@ func TestToolRegistry_ExecuteWithContext_AsyncCallback(t *testing.T) { cb := func(_ context.Context, _ *ToolResult) { called = true } result := r.ExecuteWithContext(context.Background(), "async_tool", nil, "", "", cb) - if at.cb == nil { - t.Error("expected SetCallback to have been called") + if at.lastCB == nil { + t.Error("expected ExecuteAsync to have received a callback") } if !result.Async { t.Error("expected async result") } - at.cb(context.Background(), SilentResult("done")) + at.lastCB(context.Background(), SilentResult("done")) if !called { t.Error("expected callback to be invoked") } diff --git a/pkg/tools/spawn.go b/pkg/tools/spawn.go index 8b166b41f..be40ffda2 100644 --- a/pkg/tools/spawn.go +++ b/pkg/tools/spawn.go @@ -8,25 +8,18 @@ import ( type SpawnTool struct { manager *SubagentManager - originChannel string - originChatID string allowlistCheck func(targetAgentID string) bool - callback AsyncCallback // For async completion notification } +// Compile-time check: SpawnTool implements AsyncExecutor. +var _ AsyncExecutor = (*SpawnTool)(nil) + func NewSpawnTool(manager *SubagentManager) *SpawnTool { return &SpawnTool{ - manager: manager, - originChannel: "cli", - originChatID: "direct", + manager: manager, } } -// SetCallback implements AsyncTool interface for async completion notification -func (t *SpawnTool) SetCallback(cb AsyncCallback) { - t.callback = cb -} - func (t *SpawnTool) Name() string { return "spawn" } @@ -56,16 +49,21 @@ func (t *SpawnTool) Parameters() map[string]any { } } -func (t *SpawnTool) SetContext(channel, chatID string) { - t.originChannel = channel - t.originChatID = chatID -} - func (t *SpawnTool) SetAllowlistChecker(check func(targetAgentID string) bool) { t.allowlistCheck = check } func (t *SpawnTool) Execute(ctx context.Context, args map[string]any) *ToolResult { + return t.execute(ctx, args, nil) +} + +// ExecuteAsync implements AsyncExecutor. The callback is passed through to the +// subagent manager as a call parameter — never stored on the SpawnTool instance. +func (t *SpawnTool) ExecuteAsync(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult { + return t.execute(ctx, args, cb) +} + +func (t *SpawnTool) execute(ctx context.Context, args map[string]any, cb AsyncCallback) *ToolResult { task, ok := args["task"].(string) if !ok || strings.TrimSpace(task) == "" { return ErrorResult("task is required and must be a non-empty string") @@ -85,8 +83,20 @@ func (t *SpawnTool) Execute(ctx context.Context, args map[string]any) *ToolResul return ErrorResult("Subagent manager not configured") } + // Read channel/chatID from context (injected by registry). + // Fall back to "cli"/"direct" for non-conversation callers (e.g., CLI, tests) + // to preserve the same defaults as the original NewSpawnTool constructor. + channel := ToolChannel(ctx) + if channel == "" { + channel = "cli" + } + chatID := ToolChatID(ctx) + if chatID == "" { + chatID = "direct" + } + // Pass callback to manager for async completion notification - result, err := t.manager.Spawn(ctx, task, label, agentID, t.originChannel, t.originChatID, t.callback) + result, err := t.manager.Spawn(ctx, task, label, agentID, channel, chatID, cb) if err != nil { return ErrorResult(fmt.Sprintf("failed to spawn subagent: %v", err)) } diff --git a/pkg/tools/subagent.go b/pkg/tools/subagent.go index 69f1a49a2..429340047 100644 --- a/pkg/tools/subagent.go +++ b/pkg/tools/subagent.go @@ -252,16 +252,12 @@ func (sm *SubagentManager) ListTasks() []*SubagentTask { // Unlike SpawnTool which runs tasks asynchronously, SubagentTool waits for completion // and returns the result directly in the ToolResult. type SubagentTool struct { - manager *SubagentManager - originChannel string - originChatID string + manager *SubagentManager } func NewSubagentTool(manager *SubagentManager) *SubagentTool { return &SubagentTool{ - manager: manager, - originChannel: "cli", - originChatID: "direct", + manager: manager, } } @@ -290,11 +286,6 @@ func (t *SubagentTool) Parameters() map[string]any { } } -func (t *SubagentTool) SetContext(channel, chatID string) { - t.originChannel = channel - t.originChatID = chatID -} - func (t *SubagentTool) Execute(ctx context.Context, args map[string]any) *ToolResult { task, ok := args["task"].(string) if !ok { @@ -341,13 +332,24 @@ func (t *SubagentTool) Execute(ctx context.Context, args map[string]any) *ToolRe } } + // Fall back to "cli"/"direct" for non-conversation callers (e.g., CLI, tests) + // to preserve the same defaults as the original NewSubagentTool constructor. + channel := ToolChannel(ctx) + if channel == "" { + channel = "cli" + } + chatID := ToolChatID(ctx) + if chatID == "" { + chatID = "direct" + } + loopResult, err := RunToolLoop(ctx, ToolLoopConfig{ Provider: sm.provider, Model: sm.defaultModel, Tools: tools, MaxIterations: maxIter, LLMOptions: llmOptions, - }, messages, t.originChannel, t.originChatID) + }, messages, channel, chatID) if err != nil { return ErrorResult(fmt.Sprintf("Subagent execution failed: %v", err)).WithError(err) } diff --git a/pkg/tools/subagent_tool_test.go b/pkg/tools/subagent_tool_test.go index 59bfdffae..a1450410a 100644 --- a/pkg/tools/subagent_tool_test.go +++ b/pkg/tools/subagent_tool_test.go @@ -50,9 +50,8 @@ func TestSubagentManager_SetLLMOptions_AppliesToRunToolLoop(t *testing.T) { manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil) manager.SetLLMOptions(2048, 0.6) tool := NewSubagentTool(manager) - tool.SetContext("cli", "direct") - ctx := context.Background() + ctx := WithToolContext(context.Background(), "cli", "direct") args := map[string]any{"task": "Do something"} result := tool.Execute(ctx, args) @@ -147,28 +146,14 @@ func TestSubagentTool_Parameters(t *testing.T) { } } -// TestSubagentTool_SetContext verifies context setting -func TestSubagentTool_SetContext(t *testing.T) { - provider := &MockLLMProvider{} - manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil) - tool := NewSubagentTool(manager) - - tool.SetContext("test-channel", "test-chat") - - // Verify context is set (we can't directly access private fields, - // but we can verify it doesn't crash) - // The actual context usage is tested in Execute tests -} - // TestSubagentTool_Execute_Success tests successful execution func TestSubagentTool_Execute_Success(t *testing.T) { provider := &MockLLMProvider{} msgBus := bus.NewMessageBus() manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus) tool := NewSubagentTool(manager) - tool.SetContext("telegram", "chat-123") - ctx := context.Background() + ctx := WithToolContext(context.Background(), "telegram", "chat-123") args := map[string]any{ "task": "Write a haiku about coding", "label": "haiku-task", @@ -297,12 +282,9 @@ func TestSubagentTool_Execute_ContextPassing(t *testing.T) { manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus) tool := NewSubagentTool(manager) - // Set context channel := "test-channel" chatID := "test-chat" - tool.SetContext(channel, chatID) - - ctx := context.Background() + ctx := WithToolContext(context.Background(), channel, chatID) args := map[string]any{ "task": "Test context passing", } diff --git a/pkg/tools/toolloop.go b/pkg/tools/toolloop.go index cdfe0d6ce..244f0d4a2 100644 --- a/pkg/tools/toolloop.go +++ b/pkg/tools/toolloop.go @@ -10,6 +10,7 @@ import ( "context" "encoding/json" "fmt" + "sync" "github.com/sipeed/picoclaw/pkg/logger" "github.com/sipeed/picoclaw/pkg/providers" @@ -121,37 +122,53 @@ func RunToolLoop( } messages = append(messages, assistantMsg) - // 7. Execute tool calls - for _, tc := range normalizedToolCalls { - argsJSON, _ := json.Marshal(tc.Arguments) - argsPreview := utils.Truncate(string(argsJSON), 200) - logger.InfoCF("toolloop", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), - map[string]any{ - "tool": tc.Name, - "iteration": iteration, - }) + // 7. Execute tool calls in parallel + type indexedResult struct { + result *ToolResult + tc providers.ToolCall + } - // Execute tool (no async callback for subagents - they run independently) - var toolResult *ToolResult - if config.Tools != nil { - toolResult = config.Tools.ExecuteWithContext(ctx, tc.Name, tc.Arguments, channel, chatID, nil) - } else { - toolResult = ErrorResult("No tools available") + results := make([]indexedResult, len(normalizedToolCalls)) + var wg sync.WaitGroup + + for i, tc := range normalizedToolCalls { + results[i].tc = tc + + wg.Add(1) + go func(idx int, tc providers.ToolCall) { + defer wg.Done() + + argsJSON, _ := json.Marshal(tc.Arguments) + argsPreview := utils.Truncate(string(argsJSON), 200) + logger.InfoCF("toolloop", fmt.Sprintf("Tool call: %s(%s)", tc.Name, argsPreview), + map[string]any{ + "tool": tc.Name, + "iteration": iteration, + }) + + var toolResult *ToolResult + if config.Tools != nil { + toolResult = config.Tools.ExecuteWithContext(ctx, tc.Name, tc.Arguments, channel, chatID, nil) + } else { + toolResult = ErrorResult("No tools available") + } + results[idx].result = toolResult + }(i, tc) + } + wg.Wait() + + // Append results in original order + for _, r := range results { + contentForLLM := r.result.ForLLM + if contentForLLM == "" && r.result.Err != nil { + contentForLLM = r.result.Err.Error() } - // Determine content for LLM - contentForLLM := toolResult.ForLLM - if contentForLLM == "" && toolResult.Err != nil { - contentForLLM = toolResult.Err.Error() - } - - // Add tool result message - toolResultMsg := providers.Message{ + messages = append(messages, providers.Message{ Role: "tool", Content: contentForLLM, - ToolCallID: tc.ID, - } - messages = append(messages, toolResultMsg) + ToolCallID: r.tc.ID, + }) } } diff --git a/pkg/tools/web.go b/pkg/tools/web.go index 15d2330ff..7b14686c9 100644 --- a/pkg/tools/web.go +++ b/pkg/tools/web.go @@ -395,6 +395,88 @@ func (p *PerplexitySearchProvider) Search(ctx context.Context, query string, cou return fmt.Sprintf("Results for: %s (via Perplexity)\n%s", query, searchResp.Choices[0].Message.Content), nil } +type GLMSearchProvider struct { + apiKey string + baseURL string + searchEngine string + proxy string + client *http.Client +} + +func (p *GLMSearchProvider) Search(ctx context.Context, query string, count int) (string, error) { + searchURL := p.baseURL + if searchURL == "" { + searchURL = "https://open.bigmodel.cn/api/paas/v4/web_search" + } + + payload := map[string]any{ + "search_query": query, + "search_engine": p.searchEngine, + "search_intent": false, + "count": count, + "content_size": "medium", + } + + bodyBytes, err := json.Marshal(payload) + if err != nil { + return "", fmt.Errorf("failed to marshal payload: %w", err) + } + + req, err := http.NewRequestWithContext(ctx, "POST", searchURL, bytes.NewReader(bodyBytes)) + if err != nil { + return "", fmt.Errorf("failed to create request: %w", err) + } + + req.Header.Set("Content-Type", "application/json") + req.Header.Set("Authorization", "Bearer "+p.apiKey) + + resp, err := p.client.Do(req) + if err != nil { + return "", fmt.Errorf("request failed: %w", err) + } + defer resp.Body.Close() + + body, err := io.ReadAll(io.LimitReader(resp.Body, 1<<20)) + if err != nil { + return "", fmt.Errorf("failed to read response: %w", err) + } + + if resp.StatusCode != http.StatusOK { + return "", fmt.Errorf("GLM Search API error (status %d): %s", resp.StatusCode, string(body)) + } + + var searchResp struct { + SearchResult []struct { + Title string `json:"title"` + Content string `json:"content"` + Link string `json:"link"` + } `json:"search_result"` + } + + if err := json.Unmarshal(body, &searchResp); err != nil { + return "", fmt.Errorf("failed to parse response: %w", err) + } + + results := searchResp.SearchResult + if len(results) == 0 { + return fmt.Sprintf("No results for: %s", query), nil + } + + var lines []string + lines = append(lines, fmt.Sprintf("Results for: %s (via GLM Search)", query)) + for i, item := range results { + if i >= count { + break + } + lines = append(lines, fmt.Sprintf("%d. %s\n %s", i+1, item.Title, item.Link)) + if item.Content != "" { + lines = append(lines, fmt.Sprintf(" %s", item.Content)) + } + } + + return strings.Join(lines, "\n"), nil +} + type WebSearchTool struct { provider SearchProvider maxResults int @@ -413,6 +495,11 @@ type WebSearchToolOptions struct { PerplexityAPIKey string PerplexityMaxResults int PerplexityEnabled bool + GLMSearchAPIKey string + GLMSearchBaseURL string + GLMSearchEngine string + GLMSearchMaxResults int + GLMSearchEnabled bool Proxy string } @@ -420,7 +507,7 @@ func NewWebSearchTool(opts WebSearchToolOptions) (*WebSearchTool, error) { var provider SearchProvider maxResults := 5 - // Priority: Perplexity > Brave > Tavily > DuckDuckGo + // Priority: Perplexity > Brave > Tavily > DuckDuckGo > GLM Search if opts.PerplexityEnabled && opts.PerplexityAPIKey != "" { client, err := createHTTPClient(opts.Proxy, perplexityTimeout) if err != nil { @@ -462,6 +549,25 @@ func NewWebSearchTool(opts WebSearchToolOptions) (*WebSearchTool, error) { if opts.DuckDuckGoMaxResults > 0 { maxResults = opts.DuckDuckGoMaxResults } + } else if opts.GLMSearchEnabled && opts.GLMSearchAPIKey != "" { + client, err := createHTTPClient(opts.Proxy, searchTimeout) + if err != nil { + return nil, fmt.Errorf("failed to create HTTP client for GLM Search: %w", err) + } + searchEngine := opts.GLMSearchEngine + if searchEngine == "" { + searchEngine = "search_std" + } + provider = &GLMSearchProvider{ + apiKey: opts.GLMSearchAPIKey, + baseURL: opts.GLMSearchBaseURL, + searchEngine: searchEngine, + proxy: opts.Proxy, + client: client, + } + if opts.GLMSearchMaxResults > 0 { + maxResults = opts.GLMSearchMaxResults + } } else { return nil, nil } diff --git a/pkg/tools/web_test.go b/pkg/tools/web_test.go index 8a8b88131..bdd30d385 100644 --- a/pkg/tools/web_test.go +++ b/pkg/tools/web_test.go @@ -681,3 +681,135 @@ func TestWebTool_TavilySearch_Success(t *testing.T) { t.Errorf("Expected 'via Tavily' in output, got: %s", result.ForUser) } } + +func TestWebTool_GLMSearch_Success(t *testing.T) { + server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + if r.Method != "POST" { + t.Errorf("Expected POST request, got %s", r.Method) + } + if r.Header.Get("Content-Type") != "application/json" { + t.Errorf("Expected Content-Type application/json, got %s", r.Header.Get("Content-Type")) + } + if r.Header.Get("Authorization") != "Bearer test-glm-key" { + t.Errorf("Expected Authorization Bearer test-glm-key, got %s", r.Header.Get("Authorization")) + } + + var payload map[string]any + json.NewDecoder(r.Body).Decode(&payload) + if payload["search_query"] != "test query" { + t.Errorf("Expected search_query 'test query', got %v", payload["search_query"]) + } + if payload["search_engine"] != "search_std" { + t.Errorf("Expected search_engine 'search_std', got %v", payload["search_engine"]) + } + + response := map[string]any{ + "id": "web-search-test", + "created": 1709568000, + "search_result": []map[string]any{ + { + "title": "Test GLM Result", + "content": "GLM search snippet", + "link": "https://example.com/glm", + "media": "Example", + "publish_date": "2026-03-04", + }, + }, + } + w.Header().Set("Content-Type", "application/json") + w.WriteHeader(http.StatusOK) + json.NewEncoder(w).Encode(response) + })) + defer server.Close() + + tool, err := NewWebSearchTool(WebSearchToolOptions{ + GLMSearchEnabled: true, + GLMSearchAPIKey: "test-glm-key", + GLMSearchBaseURL: server.URL, + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + + result := tool.Execute(context.Background(), map[string]any{ + "query": "test query", + }) + + if result.IsError { + t.Errorf("Expected success, got IsError=true: %s", result.ForLLM) + } + if !strings.Contains(result.ForUser, "Test GLM Result") { + t.Errorf("Expected 'Test GLM Result' in output, got: %s", result.ForUser) + } + if !strings.Contains(result.ForUser, "https://example.com/glm") { + t.Errorf("Expected URL in output, got: %s", result.ForUser) + } + if !strings.Contains(result.ForUser, "via GLM Search") { + t.Errorf("Expected 'via GLM Search' in output, got: %s", result.ForUser) + } +} + +func TestWebTool_GLMSearch_APIError(t *testing.T) { + server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) { + w.WriteHeader(http.StatusUnauthorized) + w.Write([]byte(`{"error":"invalid api key"}`)) + })) + defer server.Close() + + tool, err := NewWebSearchTool(WebSearchToolOptions{ + GLMSearchEnabled: true, + GLMSearchAPIKey: "bad-key", + GLMSearchBaseURL: server.URL, + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + + result := tool.Execute(context.Background(), map[string]any{ + "query": "test query", + }) + + if !result.IsError { + t.Errorf("Expected IsError=true for 401 response") + } + if !strings.Contains(result.ForLLM, "status 401") { + t.Errorf("Expected status 401 in error, got: %s", result.ForLLM) + } +} + +func TestWebTool_GLMSearch_Priority(t *testing.T) { + // GLM Search should only be selected when all other providers are disabled + tool, err := NewWebSearchTool(WebSearchToolOptions{ + DuckDuckGoEnabled: true, + DuckDuckGoMaxResults: 5, + GLMSearchEnabled: true, + GLMSearchAPIKey: "test-key", + GLMSearchBaseURL: "https://example.com", + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + + // DuckDuckGo should win over GLM Search + if _, ok := tool.provider.(*DuckDuckGoSearchProvider); !ok { + t.Errorf("Expected DuckDuckGoSearchProvider when both enabled, got %T", tool.provider) + } + + // With DuckDuckGo disabled, GLM Search should be selected + tool2, err := NewWebSearchTool(WebSearchToolOptions{ + DuckDuckGoEnabled: false, + GLMSearchEnabled: true, + GLMSearchAPIKey: "test-key", + GLMSearchBaseURL: "https://example.com", + GLMSearchEngine: "search_std", + }) + if err != nil { + t.Fatalf("NewWebSearchTool() error: %v", err) + } + if _, ok := tool2.provider.(*GLMSearchProvider); !ok { + t.Errorf("Expected GLMSearchProvider when only GLM enabled, got %T", tool2.provider) + } +} diff --git a/pkg/utils/media.go b/pkg/utils/media.go index a34889fb8..3e1c5d88e 100644 --- a/pkg/utils/media.go +++ b/pkg/utils/media.go @@ -3,6 +3,7 @@ package utils import ( "io" "net/http" + "net/url" "os" "path/filepath" "strings" @@ -52,11 +53,12 @@ type DownloadOptions struct { Timeout time.Duration ExtraHeaders map[string]string LoggerPrefix string + ProxyURL string } // DownloadFile downloads a file from URL to a local temp directory. // Returns the local file path or empty string on error. -func DownloadFile(url, filename string, opts DownloadOptions) string { +func DownloadFile(urlStr, filename string, opts DownloadOptions) string { // Set defaults if opts.Timeout == 0 { opts.Timeout = 60 * time.Second @@ -78,7 +80,7 @@ func DownloadFile(url, filename string, opts DownloadOptions) string { localPath := filepath.Join(mediaDir, uuid.New().String()[:8]+"_"+safeName) // Create HTTP request - req, err := http.NewRequest("GET", url, nil) + req, err := http.NewRequest("GET", urlStr, nil) if err != nil { logger.ErrorCF(opts.LoggerPrefix, "Failed to create download request", map[string]any{ "error": err.Error(), @@ -92,11 +94,24 @@ func DownloadFile(url, filename string, opts DownloadOptions) string { } client := &http.Client{Timeout: opts.Timeout} + if opts.ProxyURL != "" { + proxyURL, parseErr := url.Parse(opts.ProxyURL) + if parseErr != nil { + logger.ErrorCF(opts.LoggerPrefix, "Invalid proxy URL for download", map[string]any{ + "error": parseErr.Error(), + "proxy": opts.ProxyURL, + }) + return "" + } + client.Transport = &http.Transport{ + Proxy: http.ProxyURL(proxyURL), + } + } resp, err := client.Do(req) if err != nil { logger.ErrorCF(opts.LoggerPrefix, "Failed to download file", map[string]any{ "error": err.Error(), - "url": url, + "url": urlStr, }) return "" } @@ -105,7 +120,7 @@ func DownloadFile(url, filename string, opts DownloadOptions) string { if resp.StatusCode != http.StatusOK { logger.ErrorCF(opts.LoggerPrefix, "File download returned non-200 status", map[string]any{ "status": resp.StatusCode, - "url": url, + "url": urlStr, }) return "" } diff --git a/pkg/voice/transcriber.go b/pkg/voice/transcriber.go index f973e77fe..e949d7a22 100644 --- a/pkg/voice/transcriber.go +++ b/pkg/voice/transcriber.go @@ -10,12 +10,19 @@ import ( "net/http" "os" "path/filepath" + "strings" "time" + "github.com/sipeed/picoclaw/pkg/config" "github.com/sipeed/picoclaw/pkg/logger" "github.com/sipeed/picoclaw/pkg/utils" ) +type Transcriber interface { + Name() string + Transcribe(ctx context.Context, audioFilePath string) (*TranscriptionResponse, error) +} + type GroqTranscriber struct { apiKey string apiBase string @@ -152,8 +159,22 @@ func (t *GroqTranscriber) Transcribe(ctx context.Context, audioFilePath string) return &result, nil } -func (t *GroqTranscriber) IsAvailable() bool { - available := t.apiKey != "" - logger.DebugCF("voice", "Checking transcriber availability", map[string]any{"available": available}) - return available +func (t *GroqTranscriber) Name() string { + return "groq" +} + +// DetectTranscriber inspects cfg and returns the appropriate Transcriber, or +// nil if no supported transcription provider is configured. +func DetectTranscriber(cfg *config.Config) Transcriber { + // Direct Groq provider config takes priority. + if key := cfg.Providers.Groq.APIKey; key != "" { + return NewGroqTranscriber(key) + } + // Fall back to any model-list entry that uses the groq/ protocol. + for _, mc := range cfg.ModelList { + if strings.HasPrefix(mc.Model, "groq/") && mc.APIKey != "" { + return NewGroqTranscriber(mc.APIKey) + } + } + return nil } diff --git a/pkg/voice/transcriber_test.go b/pkg/voice/transcriber_test.go new file mode 100644 index 000000000..9b6add333 --- /dev/null +++ b/pkg/voice/transcriber_test.go @@ -0,0 +1,160 @@ +package voice + +import ( + "context" + "encoding/json" + "net/http" + "net/http/httptest" + "os" + "path/filepath" + "testing" + + "github.com/sipeed/picoclaw/pkg/config" +) + +// Ensure GroqTranscriber satisfies the Transcriber interface at compile time. +var _ Transcriber = (*GroqTranscriber)(nil) + +func TestGroqTranscriberName(t *testing.T) { + tr := NewGroqTranscriber("sk-test") + if got := tr.Name(); got != "groq" { + t.Errorf("Name() = %q, want %q", got, "groq") + } +} + +func TestDetectTranscriber(t *testing.T) { + tests := []struct { + name string + cfg *config.Config + wantNil bool + wantName string + }{ + { + name: "no config", + cfg: &config.Config{}, + wantNil: true, + }, + { + name: "groq provider key", + cfg: &config.Config{ + Providers: config.ProvidersConfig{ + Groq: config.ProviderConfig{APIKey: "sk-groq-direct"}, + }, + }, + wantName: "groq", + }, + { + name: "groq via model list", + cfg: &config.Config{ + ModelList: []config.ModelConfig{ + {Model: "openai/gpt-4o", APIKey: "sk-openai"}, + {Model: "groq/llama-3.3-70b", APIKey: "sk-groq-model"}, + }, + }, + wantName: "groq", + }, + { + name: "groq model list entry without key is skipped", + cfg: &config.Config{ + ModelList: []config.ModelConfig{ + {Model: "groq/llama-3.3-70b", APIKey: ""}, + }, + }, + wantNil: true, + }, + { + name: "provider key takes priority over model list", + cfg: &config.Config{ + Providers: config.ProvidersConfig{ + Groq: config.ProviderConfig{APIKey: "sk-groq-direct"}, + }, + ModelList: []config.ModelConfig{ + {Model: "groq/llama-3.3-70b", APIKey: "sk-groq-model"}, + }, + }, + wantName: "groq", + }, + } + + for _, tc := range tests { + t.Run(tc.name, func(t *testing.T) { + tr := DetectTranscriber(tc.cfg) + if tc.wantNil { + if tr != nil { + t.Errorf("DetectTranscriber() = %v, want nil", tr) + } + return + } + if tr == nil { + t.Fatal("DetectTranscriber() = nil, want non-nil") + } + if got := tr.Name(); got != tc.wantName { + t.Errorf("Name() = %q, want %q", got, tc.wantName) + } + }) + } +} + +func TestTranscribe(t *testing.T) { + // Write a minimal fake audio file so the transcriber can open and send it. + tmpDir := t.TempDir() + audioPath := filepath.Join(tmpDir, "clip.ogg") + if err := os.WriteFile(audioPath, []byte("fake-audio-data"), 0o644); err != nil { + t.Fatalf("failed to write fake audio file: %v", err) + } + + t.Run("success", func(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + if r.URL.Path != "/audio/transcriptions" { + t.Errorf("unexpected path: %s", r.URL.Path) + } + if r.Header.Get("Authorization") != "Bearer sk-test" { + t.Errorf("unexpected Authorization header: %s", r.Header.Get("Authorization")) + } + w.Header().Set("Content-Type", "application/json") + _ = json.NewEncoder(w).Encode(TranscriptionResponse{ + Text: "hello world", + Language: "en", + Duration: 1.5, + }) + })) + defer srv.Close() + + tr := NewGroqTranscriber("sk-test") + tr.apiBase = srv.URL + + resp, err := tr.Transcribe(context.Background(), audioPath) + if err != nil { + t.Fatalf("Transcribe() error: %v", err) + } + if resp.Text != "hello world" { + t.Errorf("Text = %q, want %q", resp.Text, "hello world") + } + if resp.Language != "en" { + t.Errorf("Language = %q, want %q", resp.Language, "en") + } + }) + + t.Run("api error", func(t *testing.T) { + srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { + http.Error(w, `{"error":"invalid_api_key"}`, http.StatusUnauthorized) + })) + defer srv.Close() + + tr := NewGroqTranscriber("sk-bad") + tr.apiBase = srv.URL + + _, err := tr.Transcribe(context.Background(), audioPath) + if err == nil { + t.Fatal("expected error for non-200 response, got nil") + } + }) + + t.Run("missing file", func(t *testing.T) { + tr := NewGroqTranscriber("sk-test") + _, err := tr.Transcribe(context.Background(), filepath.Join(tmpDir, "nonexistent.ogg")) + if err == nil { + t.Fatal("expected error for missing file, got nil") + } + }) +}